Management of Non-Small Cell Lung Cancer with an EGFR, ALK or ROS1 Abnormality — Zofia Piotrowska, MD, MHS

DR PIOTROWSKA: My name is Zofia Piotrowska from Harvard Medical School and Massachusetts General Hospital. And it’s really a pleasure to be here to talk about a year in review and some of the updates from 2022 in management of non-small cell lung cancer, specifically those with an EGFR, ALK or ROS1 abnormality. What we’ve tried to do in this program is summarize some of the most exciting abstracts and publications that have come out over the past year or so and how they impact our clinical management of these patients.

So this is an overview of all of the abstracts we’ll cover over the next series here. I’m going to break them down by first looking at EGFR mutant lung cancers, both early-stage disease and acquired resistance to osimertinib. We’ll then focus on some of the updates for the EGFR exon 20 insertions. And then, finally wrap up with updates for ALK and ROS1-positive lung cancer. So with that, let’s jump right in.

So first, we’re going to talk about EGFR mutations. And for the first 2 sections here, we’ll talk about the common EGFR mutations. These are the mutations in — that are exon 19 deletions and L858R point mutations. These are the most common type of EGFR mutations we see in lung cancer. And we know that these are highly sensitizing to standard EGFR inhibitors. And based upon the results of the FLAURA study, which is now several years old and I’ve shown you here on the right, osimertinib is our preferred first line therapy for patients with these classic EGFR mutations with a median progression free survival of about 19 months and a median overall survival in this study of about 39 months. However, there’s a lot of room for improvement in terms of treatment options for patients who progress on first line osimertinib. And we’ll talk about some of the updates there. And then, we’ve also seen over the past 2 years, the movement of osimertinib to the early-stage setting. In 2021, osimertinib became the first TKI approved in the adjuvant setting based on improvement in disease free survival in the ADAURA study which looked at 3 years of adjuvant osimertinib versus placebo.

So with that, let’s actually start with some updates from that adjuvant ADAURA study, 2 abstracts to discuss here. The first is the updated data from ADAURA. And then, an interesting analysis looking at the use of postoperative chemotherapy in that study.

So this year at ESMO, Dr Tsuboi and colleagues presented updated results from the ADAURA study. This is just a reminder of the study design, and you’ve seen this before. This is a Phase III randomized study that took patients with completely resected Stage IB to IIIA non-small cell lung cancers harboring EGFR exon 19 deletions or L858R. Patients who had a complete resection could get adjuvant chemotherapy per the investigator’s discretion and then were randomized to either 3 years of osimertinib or placebo. Nearly 700 patients here with a primary endpoint of disease free survival. And 2 years at ASCO 2020, we saw the preliminary analysis of this study with an improvement in disease free survival in these patients. And that, of course, led to the approval of adjuvant osimertinib here in the United States.

And this year, we saw updated results with longer follow-up. Here, you can see the primary endpoint of DFS in the patients with resected Stage II to IIIA disease on the left. You can see here, the DFS hazard ratio maintained here at 0.23 with a clear impressive improvement in disease free survival with adjuvant osimertinib. Nearly all of these patients had completed their 3 years of osimertinib. And there was some discussion when this was presented about the shape of these curves and the fact that they — we do start to get a hint, although these are small numbers, that these curves may start to come together after that 3 years of TKI therapy. We did not see any updated overall survival results presented in this particular abstract. And, of course, ultimately, I think that will be the question that everyone will want to know the answer to, is overall survival improved with adjuvant osimertinib? Are we truly curing more patients or delaying time to relapse? In this abstract, we also saw updated DFS by stage and specifically, an analysis looking at the current 8th edition staging system. And you can see here that adjuvant osimertinib had benefit for Stage IB to IIIA patients, but the greatest magnitude of benefit was in those patients with Stage IIIA and Stage II disease. Hazard ratios there were 0.22 for Stage IIIA patients and 0.33 for Stage II patients.

We also saw an interesting publication looking at the use of adjuvant chemotherapy in these patients and whether that had any impact on the benefit of adjuvant osimertinib. This is a somewhat busy table, but I’ll just highlight that about, in the overall study, about half of patients received adjuvant chemotherapy. And there was a lot of discussion as to whether the patients received appropriate adjuvant chemotherapy which, of course, does have a modest, but proven overall survival benefit in resected lung cancer. You can see here that the majority of patients with Stage II and IIIA disease did receive adjuvant chemotherapy, 71 and 80%. The rates of use of adjuvant chemo were lower in the Stage IB patients which I think is consistent with the kind of lower level of data supporting the use of adjuvant chemo in those patients. But importantly, if you look on the righthand side of the figure, you can see that there was benefit to adjuvant osimertinib over placebo in all of these patients regardless of whether they received adjuvant chemotherapy or not. And so, indeed, the use the adjuvant chemotherapy did not seem to have an impact on the benefit of osimertinib.

So I think the take-homes from these 2 abstracts are clearly that the updated results of ADAURA continue to show a sustained and impressive improvement in disease free survival for these patients. And based on these results, adjuvant osimertinib is approved as adjuvant therapy after resection for patients with non-small cell lung cancers with exon 19 deletions and L858R mutations. Importantly, DFS benefit appears to be maintained regardless of prior chemotherapy use. And I think it’s reasonable to consider osimertinib in both chemo eligible and ineligible patients. But in my own practice, I’m careful not to kind of substitute osimertinib for chemotherapy. And I do think that it is still really important to consider adjuvant chemotherapy for these patients, even if they may go on to osimertinib afterwards given the proven survival benefit of adjuvant chemo. Of course, the overall survival results from ADAURA will be very interesting and are eagerly awaited. And there’s some discussion ongoing about the shape of those curves and whether perhaps they may start to come together with longer follow-up as patients stop those 3 years of TKI therapy and whether future studies maybe should look at the duration — the optimal duration of TKI therapy and whether perhaps it should be even longer than 3 years. We don’t know the answer to that question. But hopefully, in coming years, we’ll get some more data there.

Moving next to the acquired resistance setting, so for patients who have progressed on first line osimertinib. We saw an abstract looking at resistance mechanisms from the ELIOS study that I presented at ESMO this year. And then, I wanted to highlight 2 novel agents, patritumab deruxtecan and amivantamab, as potential treatment options for patients who progress on osimertinib.

So looking first at the ELIOS study, this was a molecular profiling study of patients treated with first line osimertinib. And this is the study design. These were patients with newly diagnosed EGFR mutant lung cancers, all of whom received osimertinib as per the standard of care. And the goal of the study was actually to look at resistance mechanisms to osimertinib through matched tissue biopsies obtained prior to treatment start and at the time of disease progression. So really, this was a study done specifically for the goal of molecular profiling. 154 patients were actually enrolled in this overall study. And we had hoped that based on this, we would get 154 matched biopsies. And before I show you the results of the study, I think it’s sobering to look at the outcomes here. Among 115 patients who have progressed on osimertinib in this cohort, only 46 evaluable paired biopsies were able to be obtained. So about 40% of these patients, even in this trial that specifically enrolled patients highly motivated to undergo these biopsies, were actually able to undergo biopsy at the time of progression and have eval — an evaluable pre and post-treatment biopsy pair. The most common reasons that patients did not undergo a biopsy were either patient or anatomic factors that limited biopsy feasibility or technical failure of the NGS on either the baseline or progression biopsy. And, again, I think this is really sobering. We talk a lot about getting biopsies for patients who progress on first line osimertinib, but this data that suggests that this is really not so easy to do even in the context of a clinical trial, and I think really highlights the role of liquid biopsies and the need for more comprehensive methods of doing liquid biopsy analyses to try to look at some of these resistance mechanisms.

So what did we learn from those 46 paired biopsies in ELIOS? This is a chart showing the different resistance mechanisms that were seen. You can see the prevalence here on the right side of this figure. And what we saw was that overall, these findings were consistent with prior analyses where the 2 most common resistance mechanisms that were seen were the secondary EGFR C797S mutation seen in 15% of these patients and a MET amplification, a bypass pathway activation that was seen in 17% of these patients. These are important because both of these are potentially targetable resistance mechanisms either through clinical trials or in the case of MET potentially with offline EGFR and MET inhibitor — off-label EGFR and MET inhibitor combinations. What the key — another key point about ELIOS is, unfortunately, histologic transformation was not evaluated as part of this analysis although it is going to be analyzed going forward. We know from prior series that histologic transformations occur in about 15% of patients who progress on first line osimertinib. And so that’s going to be an important thing for us to look at in the future. But we also saw that many of these patients did not have a targetable resistance mechanism. And so while we think it is very important to do these biopsies or at least to try to do these biopsies when we can, what we know is that many of these patients will not have a potentially targetable resistance mechanism and need kind of broader treatment approaches, novel agents that may not be specifically honed in on one resistance mechanism or another.

So what did we learn from ELIOS? Again, I think the most important take-home here is that we talk so much about tissue biopsies, but they’re hard to obtain even in large academic centers and even in the context of prospective clinical trials. I do think it’s still important to try to get these biopsies to select treatment for these patients, but I think the onus is on us as a field to develop better liquid biopsy-based methods. Again, we saw MET amplification and C797S as the most common resistance mechanisms. And I look forward to seeing data about the rates of histologic transformation here. And going forward, I think the question will be, again, is there other methods that we could use instead of biopsies to evaluate resistance mechanisms here?

DR LOVE: So I take it that you did not do liquid biopsies on these patients which seems a shame. But if you —

DR PIOTROWSKA: They —

DR LOVE: Go ahead.

DR PIOTROWSKA: They did actually do them. This was presented just at ESMO Asia recently. So they’re looking now at the concordance of liquid and tissue biopsies there.

DR LOVE: Oh, that’s great. I was going to ask you, what would you expect? What fraction of these do you think you would pick up on liquid? Or are you actually going to have the data? Maybe you can tell me. Also, I want to ask you about, there was this NKX2-1 thing that I hadn’t seen before. What’s that?

DR PIOTROWSKA: So that was a novel finding. NKX2-1 is the gene that encodes TTF-1 which is an immunohistochemical marker that we use in lung cancer. The role of this amplification in resistance has not been previously described. So I think that now, the preclinical team is looking as to whether this actually plays a role in the development of resistance.

And to your earlier point, Neil, I think we’ll have more data about the matched tissue and plasma, but what we anticipate is that we know liquid biopsies are very good at detecting some of the secondary mutations like C797S. But what we’ve seen is that they’re less good at detecting amplification and fusion events. And then, of course, right now, the key thing that we miss by liquid biopsy are the histologic transformations. So I think hopefully, in the future, maybe we’ll have different liquid biopsy-based assays looking beyond ctDNA that may help us look at that, but we don’t yet.

DR LOVE: Yeah, I was going to ask you whether or not there’s any clues that you can get from a liquid biopsy about transformation. Like nothing?

DR PIOTROWSKA: So what we know is that we can pick up, there’s a triple mutation, so an EGFR mutation with co-mutations in p53 and RB1. Those patients, we know that if at baseline, they have those 3 mutations, they have a much higher risk of transformation. And, of course, those mutations we can pick up in the plasma. But the key thing is not all of those patients who have those triple mutations have had a transformation or will have a transformation. And so you can pick up those mutations, but you can’t right now tell whether a transformation has occurred. Because ctDNA, we’re really just looking at the genomics of these cancers. We’re not looking at protein, you know, gene expression profiles or anything else. And I think that there’s a number of novel assays trying to look at — beyond just the DNA. And I think maybe in the future, we'll be able to develop assays that will actually detect histologic transformation.

DR LOVE: I guess too, when I think about small cell, I think a lot of paraneoplastic syndrome. So I don’t know whether there’s a lot of weird stuff floating around that you could pick up.

DR PIOTROWSKA: Yeah. I think this whole area is definitely something that a lot of investigators are looking at and trying to develop those tools to see if we can find some marker of transformation that we could detect in the plasma.

DR LOVE: It’s weird. Sometimes, I see analogies. I don’t even have — I have no idea whether people in general medical oncology do also. But there’s kind of a similar situation in chronic lymphocytic leukemia and follicular lymphoma with transformation that occurs to diffuse large B-cell. I don’t know if you round on those patients anymore.

DR PIOTROWSKA: Richter’s transformation, right?

DR LOVE: Yeah, Richter’s transformation is the one from CLL — I think that’s the CLL one. And there’s the same thing — but it’s the same kind of thing, that they, you know, it’s hard to figure out without tissue and yet, they can’t always get tissue. But anyhow. Yeah. So they’re also trying to figure out is there a way that they can pick it up other than — right now, they look at scans and stuff. Anyhow. Whenever you’re ready, you can —

DR PIOTROWSKA: It’s — yeah.

DR LOVE: Go ahead.

DR PIOTROWSKA: It’s hard because I think we’ve gotten so spoiled by liquid biopsies and they’re great in some ways and we want to use them for everything. And I think we’re starting to see now the limitations of what we’re able to look at with the current assays.

DR LOVE: Right. Actually, now that I think about it, you see the same thing in prostate cancer, small cell transformation.

DR PIOTROWSKA: Yeah. Right. Exactly.

DR LOVE: Same thing as — yeah. Interesting. Okay, please continue.

DR PIOTROWSKA: All right. So looking next at some of the novel agents in development that I’m most excited about in the context of resist — osimertinib resistance in EGFR mutant lung cancer. The first to highlight is this agent, patritumab deruxtecan. This is an antibody drug conjugate targeting HER3. HER3 is overexpressed on lung adenocarcinomas and specifically enriched in EGFR mutant lung cancers. And this year, we saw the publication of the Phase I experience with patritumab deruxtecan in EGFR mutant lung cancers. As you can see here, including a cohort of 44 patients who’d had prior osimertinib and prior platinum-based chemotherapy. And here in this population with patritumab monotherapy, we saw an impressive response rate of 39%, median progression free survival of 8.2 months. So certainly, on par, if not better than some of the single agent chemotherapy options that we have available currently for these patients. And I think what’s really intriguing is if you look across the bottom of this waterfall plot here, you can see that responses to patritumab were seen in patients with a diverse spectrum of resistance mechanisms. So this agent seems to work, not just for one type of resistance mechanism for another, but more broadly. And I think this is really a promising agent for these patients.

In terms of toxicities, patritumab is essentially a chemotherapy bound to an antibody. And so many of the toxicities here shown in the high-grade. Grade 3 or higher toxicities that were seen are hematologic toxicities, as you might expect, including thrombocytopenia, neutropenia and anemia as well as some fatigue. You can see that dose reduction was required in 21% of these patients. But generally, in my experience with this agent so far, it’s been fairly well tolerated.

So patritumab deruxtecan is active in TKI and chemotherapy resistant EGFR mutant lung cancers. It’s not yet approved although it does have FDA breakthrough therapy designation here. And clinical trials are ongoing including the combination of osimertinib with patritumab in the first and second line setting. And I think a key question with this and all of the novel agents in development will be, can we find a biomarker of patritumab response? Unfortunately, in the early data, we see that IHC of HER3 expression does not seem to correlate with patritumab response. And so I think we’ll have to look at other markers. But right now, we don’t have a good biomarker to predict which patients will be in that 40% who are likely to respond.

Another combination that I’m excited about is the combination of amivantamab together with lazertinib. Amivantamab is the bispecific antibody targeting EGFR and MET which is approved, and we’ll review again in the exon 20 section of this talk. And here, in the — for the classic EGFR mutations, it’s being looked at in combination with lazertinib, a third generation EGFR inhibitor very similar to osimertinib. You can see in this cohort of patients who have been fairly pretreated, so these were EGFR mutant, classic EGFR mutations, who’d had prior osimertinib, prior chemotherapy and in some cases, other lines of therapy. We can see response rates between 25 and 30% here, even higher by blinded review in the heavily pretreated cohort, 39%. You can see that the median duration of response here was 9.6 months, median PFS was 5 months. And in terms of toxicities, we do see dermatologic toxicities with combination — with this combination, rash, paronychia, some stomatitis as well as MET-related toxicities like hypoalbuminemia and anemia — and edema. We also see that about two-thirds of these patients treated with amivantamab do have infusion-related reactions. These typically occur on cycle 1 day 1 and generally do not recur with subsequent treatments.

This is from last year’s ASCO, but I just wanted to highlight in terms of thinking about biomarkers to select patients for this combination. There was some intriguing data shown looking at this combination in patients who had identified EGFR or MET-based resistance mechanisms. Small numbers here, but you can see the response rate was even higher, 47%. Whereas in patients without an identified EGFR or MET-based resistance mechanism, the response rate was 29%. This is still early days. There was actually not any updated data about these biomarkers presented at this year’s ASCO presentation. But I think this makes sense with the mechanism of action of amivantamab as a bispecific antibody targeting EGFR and MET. And I hope that we’ll see more data on this in the future.

So amivantamab/lazertinib, I think, is another active combination here. There’s a number of clinical trials with this combination ongoing including novel delivery methods of amivantamab. And, again, here, I think we’ll need biomarkers to select the patients who are most likely to benefit, especially as we start to think about a world where maybe we have both patritumab and amivantamab and lazertinib and are trying to select patients, you know, to treat with each of these agents.

DR LOVE: Just a little bit of a diverting question out. Does amivantamab/lazertinib work, in general, in MET overexpressed tumors? And, in general, is there anything that works in MET overexpressed as opposed to mutant?

DR PIOTROWSKA: In MET overexpressed EG — and EGFR or in general like wild type?

DR LOVE: Either way. Either in EGFR or to start with. Because I know there used to be — there was an antibody that was looking at that a long time ago. I can’t remember the name of it. And just I’m not clear right now. I don’t think the new drugs work against overexpressed, do they?

DR PIOTROWSKA: So very little data here. At ASCO this year, they presented some data with amivantamab in MET exon 14 skipping mutant tumors, but I don’t think that that included MET overexpression patients. Not specifically with amivantamab, but in EGFR mutant cancers that have MET protein overexpression, there is some data that high levels of MET protein overexpression can respond to a MET TKI in combination with an EGFR TKI. But to my knowledge, I don’t think amivantamab has specifically been looked at in those patients.

DR LOVE: And just in general, if you have a patient untreated who just has MET overexpression, how do you approach those patients?

DR PIOTROWSKA: Generally, just with standard chemotherapy and immunotherapy combinations that we would a wild type patient. If we know they have high level MET expression, which I will say we don’t usually test for in our clinic, so we don’t always know about those patients in our practice, but if they did have it, if I was running out of treatment options, I may think about trying to apply for off-label approval of a drug like amivantamab just knowing its mechanism of action. But I think the data is really limited here.

DR LOVE: Okay, please continue.

DR PIOTROWSKA: All right. Okay. Moving on to our next section which is that of EGFR exon 20 insertions in non-small cell lung cancer. First, just a reminder that this is a unique kind of subgroup of EGFR mutations in lung cancer. This is quite a heterogeneous group of insertion mutations spanning exon 20, as you can see here. We do have 2 approved therapies for these patients now, mobocertinib an oral EGFR inhibitor and amivantamab an intravenous bispecific antibody targeting EGFR and MET that we just talked about in combination with lazertinib for the common EGFR mutations. Importantly, both of these are approved in the second line post-platinum setting. And this year, we saw some updated data with both of these agents as well as some of the novel agents in development.

So we’re going to talk about updates with mobocertinib and amivantamab, and then, again, abstracts looking at some new agents in development here. So first, we saw updated results from the PPP, the prior platinum treated cohort of patients treated with mobocertinib. This was the 114-patient pooled cohort from the Phase I and EXCLAIM trials. All of these patients had had at least one prior line of therapy. And you can see the results here. Essentially, the bottom line is that these results show that with longer follow-up, we see similar outcomes to what was seen in the original publication with mobocertinib with a response rate of 28% by blinded independent review and 35% by investigator review. You can see that the median progression free survival with mobocertinib in these patients was 7.3 months. But in the swimmer’s plot right in the middle of the figure here, you can see if you look at the time intervals here, quite a few of these patients have been able to remain on treatment for a year or even longer. So I think we do see some patients who reap more durable benefit from mobocertinib. And that’s consistent with median duration of response which has been, if you look across these numbers, between 15 and 17 months by blinded review. I think it is important to keep in mind that even though mobocertinib is an oral EGFR inhibitor and it feels very easy to just put a patient on an oral therapy and see them back a month or 2 later, you do have to monitor these patients carefully for toxicities, and specifically GI toxicities. We see diarrhea in 92% of patients treated with mobocertinib including Grade 3 or higher diarrhea in 23%. So it is really important to counsel patients and to monitor closely for anti — for diarrhea to institute antidiarrheal therapy if these occur. You can also see dermatologic toxicities, rash and paronychia were seen as well and other toxicities as shown on the right here.

But overall, I think mobocertinib is a good option for our patients. This updated data seems consistent with earlier reports and I think it continues to support the use of mobocertinib after chemotherapy. The rates of diarrhea here, again, are quite high and you have to watch for that. And then, I think a key question that comes up frequently now in our clinic is for that exon 20 insertion-positive patient who has progressed on chemotherapy, how do we choose between mobocertinib and amivantamab? I’ll show you the data for amivantamab next. But in my practice, I think this is really a decision that I make on a case-by-case basis together with the patient based on the different routes of administration and safety profiles of these drugs. And my goal for most patients is that they may actually be able to access both of these therapies during their disease course. Mobocertinib may eventually move to the frontline setting. There’s an ongoing Phase III study of mobocertinib versus chemotherapy first line. And then, we’ll come back to some of the novel agents which we hope will improve efficacy and safety and importantly, hopefully improve upon CNS penetration. We’ve really seen very little CNS data with either mobocertinib or amivantamab. And that’s an area of urgent unmet need for these patients.

So what about amivantamab? Again, we already covered its mechanism of action, bispecific antibody targeting EGFR and MET. You can see here that in addition to receptor degradation and inhibition of ligand binding, this drug does also have some immune cell directing activity and it’s that maybe that contributes to some of its benefit here. In this 81-patient cohort of patients with EGFR exon 20 insertions who have progressed on prior platinum chemotherapy, amivantamab monotherapy led to a response rate of 40%, median duration of response here was 11 months, and PFS was 8.3 months. You can see that the majority of patients had tumor shrinkage as you can see in the waterfall plot on the right.

In terms of toxicities, again, as we already mentioned, I think the 2 main ones to be aware of here are dermatologic toxicities including rash in 86% of these patients although largely low-grade, and infusion-related reactions. And these, as I’ve shown you in the cartoon here on the right, tend to occur most commonly on cycle 1 day 1. And there’s directions in the package insert to mitigate this with pre-dose steroids given on cycle 1 day 1 and day 2 and split dose infusions between day 1 and day 2. And with that, even the patients who do have a reaction on cycle 1 day 1, which is about two-thirds of patients, they can go on to be treated subsequently without any further reactions.

So amivantamab is another active therapy for these patients with a response rate of about 40%. Again, we talked about the key toxicities including rash and infusion reactions. This is an IV therapy as opposed to oral mobocertinib. Different safety profiles. And I think the selection between these patients has to be made on a case-by-case basis. Amivantamab is also being tested frontline, here in combination with chemotherapy in the PAPILLON study. And we hope, again, that there will be more active and better tolerated therapy options for these patients although I think amivantamab among the current drugs in development remains unique in terms of its mechanism of action. As you’ll see, many of the other drugs that are being developed are novel TKI and so even if those are successful and become approved, I think amivantamab remains unique as the only antibody-based therapy that’s approved for these patients right now.

So what about the novel agents? I wanted to highlight 2 of them. The first is this one, CLN-081 or TAS6417. This is an oral EGFR inhibitor developed to be more selective for the EGFR exon 20 insertion mutations specifically and hopefully to have a better therapeutic window and, therefore, a better safety profile and better efficacy. This is still in Phase I testing, but Dr Yu presented an update from this study at ASCO this year. You can see across dose levels, a confirmed response rate of 38%, median duration of response of 10 months. And if we look at the rates of toxicities in the overall population here, you can see that 80% of these patients did have rash, but only 1% Grade 3 or higher. And we saw that only 30% of these patients experienced diarrhea with only 3% Grade 3 or higher. Certainly, in terms of GI toxicities, I think we are starting to see improvement with this agent.

There was also updated data presented this year with sunvozertinib. And here, we saw a pooled analysis of all the patients treated across the ongoing studies of this agent treated at 300 mg daily. In this 84-patient cohort, the combined response rate with sunvozertinib was 52%. And in the ongoing WU-KONG6 trial, which is a Phase II study being done in China, the objective response rate with 300 mg daily of sunvozertinib was 59.8%. So I think this is really promising. We’re starting to see an improvement in the response rates that we’re seeing here. And you can see the toxicities, which I’ve kind of focused on the ones at 300 mg, diarrhea in 59% of these patients, rash in 39%. So not toxicity-free, but hopefully, we are starting to see kind of a decrease in the rates of some of these toxicities, and an improvement in the efficacy profile.

So I think both of these drugs are showing promise, both in terms of efficacy and safety. We did see anecdotal CNS responses with both of these agents reported in these abstracts, however, we don’t have dedicated CNS cohorts to evaluate their CNS efficacy. And importantly, I’ll say that there is a couple of new drugs that are now entering Phase I studies that are predicted preclinically to have CNS penetration, BLU451 and ORIC114. And I think it’ll be interesting to see what those data show.

DR LOVE: Are these agents basically similar to mobocertinib?

DR PIOTROWSKA: Yeah. I think there are subtle structural differences that make them hopefully be more selective. But, in general, yes. They’re all oral therapies. They’re similar to mobocertinib. In many ways, I think they’re also similar to osimertinib. But I think with the structural changes that we see with exon 20 insertions, even small differences in how these drugs are developed can lead to better inhibition and a better therapeutic window.

DR LOVE: Is the thinking that you would have greater efficacy or more likely better tolerability?

DR PIOTROWSKA: I think the hope is that the 2 will go hand-in-hand, that if you have a drug with a wider therapeutic window, if you can dose it to higher levels against the exon 20 insertions without wild type EGFR inhibition that you’ll have better efficacy and better safety.

DR LOVE: Okay, please continue.

DR PIOTROWSKA: Great. So in the final section, we’ll talk about ALK and ROS1 alterations in non-small cell lung cancer. For ALK-positive disease, we actually have a wealth of ALK inhibitors approved in the United States. You can see them all listed here including the first generation ALK inhibitor, crizotinib, as well as the second generation ALK inhibitors, ceritinib, alectinib and brigatinib and lorlatinib, all approved for frontline use in ALK-positive disease. And I think the key question in ALK-positive disease in recent years has been which is the best agent to use upfront for these patients? We’ve had randomized Phase III studies of the second generation ALK inhibitors, alectinib and brigatinib, and recently, the third generation ALK inhibitor, lorlatinib, all compared to crizotinib in the frontline setting. All of these were positive studies. And I’ll show you some updates from these trials here. And then, for ROS1-positive disease, we have 2 approved agents, crizotinib and entrectinib. And I’ll show you some updated data with entrectinib as well as a new drug to keep an eye on for ROS1.

So first, ALK-positive disease. Let’s first talk about the updates from the Phase III J-ALEX study and the ALTA-1L trial. And then, we’ll go over some updated data from the lorlatinib first line study, the CROWN trial.

So J-ALEX, this was a Phase III study done looking at alectinib versus crizotinib in a Japanese population. This was the first study of its kind done looking at a second generation ALK inhibitor compared to crizotinib. We saw previously, improvement in progression free survival with alectinib in this study. And this year, we saw the publication of the final OS results from the study with more than 5 years of follow-up. I think probably not surprisingly, alectinib did not improve overall survival with crizotinib. The hazard ratio here was 1.03. But what we saw was really impressively high 5-year survival rates in both arms of the study, 60 and 64% of these patients alive at 5 years. I think really remarkable numbers. And what this suggests is that these results were likely confounded by the high rates of crossover. Nearly 80% of the patients who originally received crizotinib were able to get alectinib in later lines of therapy. And I think also, access to lorlatinib and other novel ALK inhibitors will really washout the effects of that first line therapy on overall survival.

We also saw the updated final results from ALTA-1L. This was the Phase III study of first line brigatinib, the second generation ALK inhibitor, versus crizotinib. Again, previously, we had seen a clear improvement in progression free survival with brigatinib. And then, these final results with a median follow-up of over 40 months in the brigatinib arm, we saw clearly continued benefit with first line brigatinib when compared to crizotinib. The final PFS hazard ratio here was 0.48 with a 3-year landmark PFS of 43% of patients progression free at 3 years with brigatinib, keep that number in mind, 43%, when we come back to the lorlatinib trial, and 19% with crizotinib. And, again, here, like in J-ALEX, there was no overall survival improvement in the overall population, probably not surprisingly. But interestingly, the post-hoc analysis showed a hint of an overall survival benefit in those patients with baseline brain metastases. And I think one of the key distinguishing factors of the second and third generation ALK inhibitors over crizotinib has been their improved CNS penetration. And I think this suggests that for those patients with baseline brain mets, if you can treat with a more CNS penetrant agent early, you can see real clear improvements in outcomes.

So I think both of these studies have continued to show benefit to the second generation ALK inhibitors over crizotinib. Not surprising. We expected these data. Again, I’m not surprised that there are no overall survival differences in these studies of first line therapy. One, we have so many treatment options for patients with ALK-positive lung cancer. And when patients with ALK-positive lung cancer are living so many years, again, with more than 60% of patients alive at 5 years, it’s really remarkable to see. I think both of these agents remain good first line treatment options for our patients. But the real question is how do we select between these and the third generation ALK inhibitor, lorlatinib, in light of the CROWN study which I’ll show you next. And I think another key question is will we ever have a first line study of a second versus a third generation ALK inhibitor? I suspect the answer may be no, but I think that would really be the question that would help inform our practice right now.

So looking at the updates from the CROWN trial. So CROWN is the Phase III study looking at first line use of lorlatinib. This is the third generation ALK inhibitor, very CNS penetrant agent with a broad activity against different ALK resistance mutations. This was here compared to first line crizotinib in ALK-positive lung cancer. We saw the publication of an unplanned interim analysis with a median follow-up of 37 months in the lorlatinib arm. And here, you can see that the progression free survival in this — hazard ratio in this interim analysis was 0.27. And if we look at that landmark of 3 years here in this study, you can see that 64% of the patients on the lorlatinib arm remained progression free at 3 years. And that does, looking across trials with all of the flaws of cross-trial comparisons, compared favorably with what we saw on both the ALEX and ALTA-1L studies with alectinib and brigatinib where the 3-year progression free survivals were in the 40, 44 — 40 to 45% range. We also saw that response rates to both drugs were high, but intracranial response in patients with baseline measurable brain metastases was significantly higher in those patients treated with lorlatinib, 83 versus 23%. And we also saw in this analysis the time to intracranial progression was clearly longer in patients treated with lorlatinib, the much more CNS penetrant therapy, than crizotinib, as you can see on the righthand Kaplan-Meier curve here. The hazard ratio here was 0.08. So clearly, patients do better in terms of CNS outcomes with lorlatinib.

We also saw a publication specifically looking at CNS outcomes in the CROWN trial, both in terms of subgroup analyses of patients with and without baseline brain metastases, and I think some interesting data on the management and types of CNS adverse events that were seen. So first, looking at the outcomes. You can see here that progression free survival was improved with lorlatinib, both in patients with and without baseline brain metastases, as you can see on the left here. And impressively, if you look at patients with CNS metastases at baseline, you can see complete CNS responses were seen in 61% of patients treated with lorlatinib. This included both the measurable — patients with measurable brain metastases which are shown in the waterfall plot here, but also patients with non-measurable brain metastases. This was significantly higher than what we saw with crizotinib, again, a drug that we know does not have very good CNS activity.

One of the, I think, concerns about the CROWN data and the use of first line lorlatinib has been the toxicity profile of this agent and specifically, some of the CNS adverse events that could be seen with lorlatinib and how that might impact patient quality-of-life when they’re getting first line lorlatinib, particularly if they’re on it for many years. Here, they looked at the incidence and the types of adverse events that were seen — CNS adverse events that were seen with lorlatinib. And what we saw in this analysis was that about 35% of these patients had CNS adverse events overall, the majority of these were Grade 1 and 2, and that cognitive and mood effects were the most common. And you can see the specific types of cognitive and mood effects that have been seen. We saw that speech effects and psychotic events — effects were quite uncommon, single digits here, and no high-grade speech or psychotic effects seen. I think what we saw here was that the median time to the first episode of CNS adverse events was about 57 days. But what was most intriguing was that these are generally well managed with dose modifications of lorlatinib, either dose interruption or sometimes, dose reduction. And what I found most intriguing was that landmark progression free survival analysis showed a comparable PFS efficacy in patients who required dose reduction than those without. And I think in a first line study, we always worry, well if we put patients on lorlatinib, but then we have to reduce the dose because of some of these toxicities, will they still benefit the way that we saw in this study? And this data, I think, gives us some comfort to feel okay interrupting or dose reducing lorlatinib for CNS adverse events. And I think clinically, that’s very helpful data.

So I think continued follow-up for the CROWN study and the CNS-specific analyses really continue to show impressive results with first line lorlatinib. 64% of these patients remain progression free at 3 years. It’s really remarkable numbers. I think these results looking across clinical trials do compare favorably with the data seen with the second generation ALK inhibitors. We clearly see that intracranial response rates are substantially higher with lorlatinib than crizotinib. And lorlatinib does appear to delay time to CNS progression. CNS adverse events are a problem with this agent, certainly. We see them in about one-third of patients overall. But they can be well managed with dose modification and that does not seem to adversely affect PFS. So I think based on these data, lorlatinib is now a preferred first line TKI for newly diagnosed ALK-positive lung cancer. And in my practice, it’s definitely my preferred agent for those patients who are diagnosed and have baseline brain metastases at diagnosis. Lots more questions, I think, to answer including a better understanding of resistance mechanisms to first line lorlatinib, how we can think about maybe high-risk subgroups including specific ALK variants or comutations. But for now, I think the CROWN trial data really have — are starting to change our practice.

So in the final section, let’s talk — oh, sure. Go ahead.

DR LOVE: Hold on. Yeah. I’ve got a million questions there, but just, you know, we don’t have to get into all of it, but just a few things. I love that term, a preferred option. Up until recently, I think it was actually ASCO, the only person I could find who was saying they gave first line lorlatinib was your colleague, Dr Gainor. But it sounds like you’re starting to move in that direction. What about without brain mets?

DR PIOTROWSKA: I think over this year, I will say, my own view on this has changed. I think with longer follow-up from lorlatinib, I’m convinced that this data really does look better and I think if we can manage these CNS adverse events with dose reduction and patients don’t have adverse impacts on efficacy, to me, that’s enough to start to change my practice to use lorlatinib for all patients with ALK-positive disease.

DR LOVE: Well that was my other question which is, why not do sort of dose escalation, you know, start out low?

DR PIOTROWSKA: Yeah, I think we are starting to look at that. And I think a key point about lorlatinib is the approved starting dose of 100 mg daily was really approved to overcome certain ALK resistance mutations which is not needed in that frontline setting. We may not need 100 mg daily. And so often, I think we do consider starting at a slightly lower dose, something like 75 mg daily, and seeing how patients do and then maybe escalating if they’re doing okay. Conversely, I think you can start at 100 and just have a very low threshold to dose reduce. But I think the key point here is you have to watch these patients closely because these CNS adverse events, they can be bothersome to patients. And I think they do respond very well to dose interruptions and dose reductions.

DR LOVE: And what’s the time sequence of the CNS issues?

DR PIOTROWSKA: So in this analysis, the median time to onset was about 60 days. I’ve seen it even earlier than that, but it takes at least, I think, a few weeks to start to see it.

DR LOVE: How about later? Do you see people who are doing fine and then all of a sudden, they start having problems 6, 12 months later?

DR PIOTROWSKA: In my experience, that’s been less common. If they’re going to have them, they started to have them early on. And then, if anything, they get better because we figure out how to manage them with dose holds or with dose reduction. But I think you certainly can see them develop later on less commonly.

DR LOVE: Incidentally, I was looking — the reason I was kind of looking down is I was trying to actually find the hazard rate on brain mets and I didn’t — I thought it was the p-value, 0.08. It was actually the hazard rate which is amazing.

DR PIOTROWSKA: It is, isn’t it?

DR LOVE: But actually — the curves are incredible, but actually when you think about it, I know it was a totally different situation, but if I remember correctly, in ADAURA, granted it was adjuvant, I think it was 1% versus 10% brain mets.

DR PIOTROWSKA: That’s right. Yeah, that is right.

DR LOVE: So it’s kind of in a way, that’s, you know, hazard rate of 0.1 or whatever.

DR PIOTROWSKA: 9%, yeah. Yeah, you’re right.

DR LOVE: But the real — the other question I have is, I just am trying to figure out kind of the way the FDA looks at stuff. And you’re talking about ADAURA. And I know it’s adjuvant, but we’re waiting for ADAURA. And now, maybe the PFS is not looking that great. But just trying to figure out what it takes to get FDA approval. I don’t know if you know, but you probably run into your colleague, Noopur Raje, there is myeloma.

DR PIOTROWSKA: Yeah.

DR LOVE: They just had an antibody drug conjugate that had an accelerated approval with a single arm response rate of 41%. It just got pulled off the market.

DR PIOTROWSKA: Why?

DR LOVE: Because it didn’t beat a standard therapy in a randomized trial. And so, you know, whatever. It’s just another, you know, and now, I look here at the survival and I want — I know that if — yes, it’s adjuvant, but kind of, it’s still, in a way, a quality-of-life issue, I think.

DR PIOTROWSKA: Yeah, I think so.

DR LOVE: When you give lorlatinib rather than alectinib or even crizotinib, I don’t think you’re giving it to get a survival benefit. I think you’re giving it to prevent brain mets, quality-of-life, even longer PFS is — given the tolerability problems. It’s not an IV infusion with a lot of toxicity. So I don’t know. I’m just trying to figure out like what it is that takes, you know, I know accelerated approval messes things up when you get — that’s what happened with this myeloma drug.

DR PIOTROWSKA: Yeah.

DR LOVE: It had accelerated approval and then their Phase III trial didn’t pan out. So unfortunately, now the drug is not even available. But anyhow. I’m just kind of —

DR PIOTROWSKA: Well I think —

DR LOVE: Go ahead.

DR PIOTROWSKA: What I was going to say is I think the challenge with these studies, and we saw this with those overall survival curves from J-ALEX and from ALTA-1L, is when you have patients who are living more than 5 years, you know, many years, it’s really hard to be able to evaluate the overall survival impact of first line therapy, even if they’re on that first line therapy for several years. There’s so many other drugs and that effect is washed out with longer follow-up. And so I think OS is a really difficult endpoint to assess here.

DR LOVE: Yeah. And, again, I hate to bring up these analogies other than I know oncologists are thinking about it.

DR PIOTROWSKA: Yeah.

DR LOVE: But I think about CLL. And now, CLL, the average PFS is 5 years, same thing as what you’re saying.

DR PIOTROWSKA: Yeah.

DR LOVE: So you can’t see a survival benefit because, again, the treatment of relapse has great effect also.

DR PIOTROWSKA: Exactly.

DR LOVE: So I guess that this is going — I guess that the reality is these kinds of situations, you’re really thinking about, I guess, quality of life.

DR PIOTROWSKA: I think so. I think certainly, in terms of our selection of these drugs, I think even a lot of the debate that you’ve seen about whether we should use alectinib or lorlatinib, it really has hinged on quality-of-life. If lorlatinib, I think, didn’t have some of these unique CNS adverse events, I think there wouldn’t have even been this debate over the past couple of years. But when you have patients who are living for 3 years+ on these therapies, it becomes a very important part of the equation.

DR LOVE: And it’s so challenging when you don’t have direct randomized data.

DR PIOTROWSKA: Right.

DR LOVE: But that happens all the time in oncology, all the time. All right.

DR PIOTROWSKA: And I don’t know that we will have a randomized study of lorlatinib versus alectinib, for example. I don’t know that that’s ever going to happen.

DR LOVE: Well just to give you another insight into the kind of content that’s getting filtered to general medical oncologists, 2 days after that webinar we’re doing with you and Greg, we’re doing a 2-and-a-half-hour program with 4 people on just ALK.

DR PIOTROWSKA: It’s great. I think it’s great that there’s so much interest and people want to hear about it for what I presume is a disease that they have maybe 1 or 2 patients in their practice with.

DR LOVE: I know. It’s amazing. You talk to these people, you’re right, they have like 1 or 2 people in their — but they still want to know and they really need to know. You know? All right.

DR PIOTROWSKA: Okay. In the final section, we’re going to briefly talk about ROS1-positive lung cancer. Again, here, we currently have 2 approved agents, crizotinib and entrectinib. We saw some updated data with entrectinib that I’ll show you. And then, I wanted to highlight a novel agent, NVL-520 which I think is one to keep an eye on in the setting of ROS1 TKI resistance.

So first, updated data from the Phase I/II studies of entrectinib. This was presented by Dr Fan and colleagues at the World Lung Cancer Conference this summer. Here, we saw updated data with entrectinib in patients with ROS1. We continue to see really positive results here with entrectinib. You can see the overall response rate here was 67% in this overall efficacy population and also in the cohort of patients who received entrectinib first line. The median duration of response here was 20 months in the overall population, 35.6 months in that first line population. You can see the progression free survivals here along the bottom of the curve. One key distinguishing factor of entrectinib versus crizotinib is entrectinib’s superior CNS activity. We saw in this cohort that the intracranial response rate of entrectinib was 49% overall and 69% in the patients treated in the frontline setting. So for patients with CNS metastases, I think, entrectinib does have the advantage here. Entrectinib side effects were most commonly dysgeusia seen in 43% of these patients, weight increase seen in 38%, dizziness, constipation and diarrhea, as you can see here. Some of these are due to off-target effects against TRK. And we have seen this in previous series before.

But overall, I think this data is encouraging. It clearly shows that entrectinib’s efficacy is maintained in this analysis. And entrectinib remains a standard of care for patients with ROS1-positive lung cancer. I think the high intracranial response rates here make this my preferred agent for patients with ROS1-positive cancers and baseline brain metastases. For those patients without baseline brain metastases, I think both entrectinib and crizotinib are reasonable options. And it’s worth having a discussion about the different toxicity profiles of these agents with patients. There is actually a Phase I — a Phase III study of first line entrectinib versus crizotinib ongoing in ROS1-positive patients. And so that will be very interesting to see.

Finally, just a few weeks ago, we saw the publication of this data with — this preclinical data with NVL-520. This is a novel, rationally designed ROS1 inhibitor that’s been developed to overcome the ROS1 G2032R solvent front resistance mutation which has been historically difficult to overcome with the currently available agents. It’s also predicted to have good intracranial efficacy. And this was really developed with the clinical need for new drugs for patients who develop acquired resistance to the currently available ROS1 TKIs. And what you can see here is this publication was largely a preclinical one. I’ve just shown you one figure showing the high predicted potency of NVL-520 against ROS1 cell lines and those cell lines with ROS1 fusions and G2032R resistance mutations. And in this paper, they also noted 3 clinical responses that have been seen in patients so far including 2 patients with the G2032R resistance mutation and 1 patient who had intracranial progression on prior ROS1 inhibitors. So this is still very early data, but I think this is a drug to keep an eye on. We really haven’t had good treatment options for patients with ROS1 who progress on the standard currently available ROS1 inhibitors. And so I think based on this preclinical and early clinical data, it’s certainly worth considering referrals of these patients for Phase I trials of this agent, NVL-520. And hopefully, we’ll have more data to come about this agent over the coming year.

So with that, I’ll say thank you. It’s been fun to review the data with targeted therapies for EGFR, ALK and ROS1-positive cancers. We’ve really seen an abundance of data in both the early stage and advanced stage setting with targeted therapies in this space in 2022. And I think we’re all excited to see what 2023 will bring. Happy New Year.

DR LOVE: That was awesome. There was one question I jotted down here I forgot to ask you, sort of not exactly humorous. But I was just wondering what’s a Grade 1 hallucination?

DR PIOTROWSKA: Actually, I have to say, overall, looking at that data and from my own clinical experience, what stood out to me was the fact that these are very difficult to grade. These are such subjective things that patients are describing when they come in and they say I’m having mood effects or — it’s really difficult as an investigator to grade these and to figure out the severity. So I don’t know what that means.

DR LOVE: That was the other thing I was thinking about also was, you know, the sort of like the color, the visual changes you see with crizotinib.

DR PIOTROWSKA: Yeah.

DR LOVE: I know that’s its own thing, but in a way, why does that happen? I don’t know. Do you know why? Is there anything new on why that happens?

DR PIOTROWSKA: I don’t know. I haven’t looked recently as to see exactly why that’s happening. It does seem to be really different than what they see with lorlatinib.

DR LOVE: Yeah. I guess it’s a completely separate thing. But I also wonder, do you think that there might be, I don’t know, like subjective effects on their consciousness or some kind of — not a hallucination or delusion, but maybe something milder? That’s what I was trying to get at with this Grade 1. Is there something else that maybe they feel that we’re not picking up?

DR PIOTROWSKA: Yes. Actually, I think that’s definitely the case. I think we’ve seen mild personality changes in some patients, like family members coming in and telling us that the patient thinks they’re fine, but they’re noticing subtle differences. So I absolutely think that there’s, you know, in all of these patients, there’s probably mild effects that are just really hard to describe, for them to describe to us and to quantitate.

DR LOVE: I was wondering whether some of — I’ve heard about these personality changes and I think I’ve heard sometimes that people like the whatever. But I was wondering whether or not it’s possible it actually would be pleasant. As opposed to an adverse feeling, maybe they — it makes them feel like not high, but better or something. I don’t know.

DR PIOTROWSKA: So I have had patients who have described almost like a not quite speed effect, but kind of an activating effect that is, you know, it’s not fully un — it’s not fully pleasant, it’s not fully unpleasant. But I have had patients who’ve described that, that it does give them more energy, it gives them, it kind of rubs them up which, in some ways, is helpful. But ultimately, I think when patients have had that, most of them have said the problem is I don’t feel quite like myself or it just doesn’t feel like a natural feeling. And that’s been the part that’s been most challenging. But these can — I think the other thing is that these can vary widely. Some people, it’s mood effect. Some people, it’s memory impairment. It can be all sorts of different things and they can be really hard to kind of categorize.

DR LOVE: I wonder has anything been looked at in terms of blood levels. Could you actually — is there anything you could draw that might help you figure out whether you’re at the right dose or not?

DR PIOTROWSKA: I think we certainly saw in the Phase I studies that these were more common at the higher doses. And so I think def — I think that’s part of the decision making that went into the recommended Phase II dose of lorlatinib. And it was, these were more problematic at the higher doses and, again, respond to dose reduction. But I don’t know that we have right now a way to kind of check blood levels and figure out can we or should we make a dose adjustment.

Therapeutic Approaches for Patients with Other Actionable Genomic Alterations — Gregory J Riely, MD, PhD

DR RIELY: Thanks, Neil, for having me today. I get the opportunity to talk about some of these less common driver oncogenes, well at least the newer ones, how about that, the ones that aren’t quite the top of mind things like EGFR and ALK. And so I think we have some exciting data over the past year and I’m happy to run through some of that.

One of the big new oncogenes that we can talk about today is ERBB2, or HER2-mutant non-small cell lung cancer. For many years we’ve talked about HER2 in the context of breast cancer, where we think of it as an amplified oncogene. But in lung cancer there are HER2-mutant forms of non-small cell lung cancer. It happens in a couple percent of people with non-small cell lung cancer, and it’s been something that we’ve known about for a few years, but we haven’t had a good drug to target it, and over the last year we’ve really seen some data come to the fore that demonstrate that trastuzumab deruxtecan can have a role for the treatment of patients with HER2-mutant non-small cell lung cancer.

This is a paper from The New England Journal of Medicine that came out this year that showed that in patients who’ve had prior treatment for their non-small cell lung cancer, those patients all had HER2 mutants, the response rate was 55%. And you can see across the bottom that the thing that unites all these patients is their HER2 mutation. There’s different levels of HER2 amplification and HER2 expression. And most of the patients in this trial had not had any prior HER2 tyrosine kinase inhibitors. So this definitely tells us a 55% response rate, certainly a respectable response rate in the second-line setting for trastuzumab deruxtecan.

Importantly, this is the first antibody-drug conjugate that’s been studied in patients with lung cancer — or approved in patients with lung cancer, so potentially an exciting new path forward.

Now we always go behind response rate and try to understand how long the benefit of these treatments is. And so this is the evidence for trastuzumab deruxtecan in patients with HER2-mutant non-small cell lung cancers, and you see the median progression-free survival is about 8 months. I think the response rate and progression-free survival we see here are certainly enticing, but at the same time kind of a first-generation level of benefit for patients with HER2-mutant lung cancer, and hopefully we can push forward on this front.

So just to summarize this data set, trastuzumab deruxtecan in patients with HER2-mutant non-small cell lung cancer. I’d emphasize that identifying HER2 mutations in patients with non-small cell lung cancer gives you a targeted therapy option so this has got to be part of your testing platform. And in the setting of patients with previously treated HER2-mutant non-small cell lung cancer trastuzumab deruxtecan has reasonable efficacy as evaluated by both response rate and progression-free survival.

We have evidence for the efficacy of trastuzumab deruxtecan. There’s been a recent question about what’s the right dose of trastuzumab deruxtecan. This is data from a randomized study where patients with HER2-mutant metastatic non-small cell lung cancer who had had prior therapy were randomized to either 5.4 mg/kg or 6.4 mg/kg. Now the dose that’s FDA approved is the 5.4 mg dose, so this is testing a higher dose to see if there’s any potential benefit or any change in toxicity.

So the first thing that we care most about, of course, is efficacy. And you see here that when we compare the response rate for the 5.4 mg/kg dose to the 6.4 mg/kg dose that there’s no clear benefit in terms of toxicity. So you see on the left, for the lower dose a 54% response and for the higher dose a 43% response rate. I’m not going to say those are different, but there’s no clear signal that the higher dose is more beneficial.

In addition to efficacy we always care about toxicity. And as you’d expect the higher dose of trastuzumab deruxtecan leads to greater toxicity. This is an overview of the toxicity data rather than focusing in on any particular toxicities. But if you look at the any-grade toxicities or greater than Grade 3 toxicities, which I think is probably the one that means the most to our patients, for the higher dose we saw 58% of patients with Grade 3 or greater toxicity, on the higher dose, and on the lower dose we saw 32% of patients with Grade 3 or greater toxicity. So again, just as you’d expect in terms of toxicity. We see more toxicity for the higher dose.

So just to summarize on this question around dose of trastuzumab deruxtecan, I think both 5.4 mg/kg and 6.4 mg mg/kg have efficacy in patients with previously treated HER2-mutant non-small cell lung cancer. There was no clear increase in efficacy with the higher dose. There was a lower incidence of ILD in data I didn’t show you, as well as other adverse events with the 5.4 mg/kg dose. And in the author’s presentation of this data from ESMO 2022 they had this summary statement, which I wholeheartedly support, and says “The totality of the evidence and a compelling positive benefit/risk balance support the FDA’s approval of trastuzumab deruxtecan at 5.4 mg/kg as the first HER2-targeted treatment for patients with previously treated HER2-mutant non-small cell lung cancer and support the establishment of trastuzumab deruxtecan as the new standard of care in this patient population.”

So we focused so far on HER2 mutations. And again, HER2 mutations are different than the amplifications we talk about in breast cancer. But there’s been some data exploring the value of using HER2 overexpression, so kind of turning back the clock on HER2, and instead of using any of these fancy molecular techniques using immunohistochemistry and looking to see how much HER2 expression there is.

This is data presented at ESMO this year of a group of patients with metastatic non-small cell lung cancer who had HER2 overexpression, and in this context it’s IHC 2+ or 3+. Again, we have a little bit of information about the 2 different doses, 5.4 mg and 6.4 mg/kg, but what I’d emphasize here is that we are seeing responses in these patients with HER2 overexpression. And I’d highlight that this trial excluded — or this arm of the trial excluded patients with HER2-mutant non-small cell lung cancer. So this is really selecting only based on HER2 overexpression.

And so on the bottom you see response rate is 34% for the FDA-approved 5.4 mg/kg dose, and on the top panel, for the higher dose, the response rate is in the similar ballpark, maybe a little bit lower.

Now we’re seeing responses, which is certainly encouraging, and the median duration of response is 6 months, so again encouraging. This is not the same as what we see for first-line treatment of EGFR-mutant lung cancer, but at the same time, for this group of patients without an oncogenic-directed therapy available to them, this certainly shows some activity and something that we can potentially move forward.

Again, this is one of the first antibody-drug conjugates that’s being explored in patients with lung cancer, and I think it’s certainly an exciting arena.

So just to summarize, trastuzumab deruxtecan has efficacy in patients with HER2 overexpression, and that’s either HER2 2+ or HER2 3+.

DR LOVE: So a couple of clinical questions I’m curious about. Putting aside cost and reimbursement, do you think there’s adequate data right now — or how would you, or how are you using T-DXd both in HER2 mutant and HER2 overexpressing? I’ve heard people talk about using it first line in mutation. I don’t know whether it’s considered, for example, second line overexpresses or maybe not enough data. What’s your take right now on the clinical utility or where this should fit in?

DR RIELY: So I think it’s value in HER2-mutant patients is very clear. I think there’s real efficacy data that seems to me clearly superior to what we’ve seen with first-line platinum doublet in combination with checkpoint inhibitors. And so I think it’s a reasonable first-line approach for a patient with HER2-mutant non-small cell lung cancer.

It’s label is for second-line, and that’s a perfectly appropriate place to use it, as well, and probably where I end up using it most.

With regard to its use in patients with overexpression, I think this data is evolving. It certainly looks encouraging. It certainly looks to be active, and I think in the second-, third-line context, where we have very modest drugs available for those patients without an oncogene, this is a perfectly reasonable approach. I’ll admit to having not explored its use off label yet, but I think it is something that we have to look forward to.

And of course, this is — it raises the question of how much more molecular testing we’re going to be — or how much more testing we’re going to be able to do for our patients. This is not something that’s available on a next-generation sequencing panel and so would require immunohistochemistry on tumor biopsy to detect.

DR LOVE: So the other question I have if let’s maybe focus on HER2-mutant disease, and you reiterated something, again, I’ve heard a lot, which is the thought of using it first line. But I guess my question is, is this really targeted therapy? And do these patients fit the targeted therapy model? In terms of being targeted therapy, in a way kind of you could view this as chemotherapy, to me, I guess, as opposed to targeted therapy.

And I guess the biggest question, and I’m going to ask you the same question later on about KRAS, is what do we know about the concept of IO in HER2-mutant disease? Do you look at them like wild type, like you expect to get the same thing out of IOs and chemo? Or you look at them more like EGFR, ALK, they don’t really benefit from IO? And how does that fit into your idea of using it first line?

DR RIELY: Yeah. I think that patients with HER2 mutations are more in that category of patients like EGFR, like ALK, both in their clinical phenotype, I mean they’re often never smokers, they almost always have an adenocarcinoma, and to my mind they’re relatively lower likelihood of benefit with an immune checkpoint inhibitor. So I think this is that group of patients who we’re not getting amazing responses with single-agent checkpoint inhibitor.

DR LOVE: That’s really interesting.

DR RIELY: And as a consequence — without that evidence to say that checkpoint inhibitors lead us to great efficacy in this context, that’s what makes me think first-line use might not be unreasonable.

DR LOVE: Do you generally think about checkpoint inhibitors second line? Or would you be thinking about it?

DR RIELY: Yeah. I never say never to a checkpoint inhibitor, even in these patients with EGFR and ALK there are the handful of cases where we’ve seen tremendous responses, and we don’t want to miss that opportunity for our patients. And so most times I will try to give a line of therapy with a checkpoint inhibitor, oftentimes as a single agent in the second-, third-line context.

DR LOVE: Please continue.

DR RIELY: So we’ll move next to KRAS. KRAS G12C mutations are a common oncogene. They happen in about 10 or 12% of patients with non-small cell lung cancer, and this field has moved very quickly with the approval of 2 drugs in the last year and a half, sotorasib and adagrasib. And this — these approvals both happened based on the context of single-arm Phase II trials where we saw reasonable response rates and reasonable progression-free survival for this group of patients without another targeted therapy option.

But of course the gold standard is a randomized controlled trial. And what you see here are the results of a randomized Phase III trial comparing sotorasib to conventional second-line docetaxel. Now I highlight this isn’t docetaxel plus ramucirumab, which has been shown to be superior to docetaxel alone. This is just single-agent docetaxel given on that 75 mg/m2 q3wk schedule that I think relatively few of us use.

But patients were randomized. All patients had KRAS G12C, and they were randomized to either sotorasib or docetaxel. So the primary endpoint of the trial was progression-free survival, and you see that here on the top panel. And there was an improvement in progression-free survival, hazard ratio of 0.66. Small p-value. Certainly an encouraging benefit from sotorasib compared to docetaxel.

What was a bit striking, and I think what surprised most folks is that based on this PFS I think we would have expected a difference in overall survival, but there was absolutely no difference in overall survival. You see the hazard ratio’s 1.01, p-value of 0.53, and the median overall survival for both sotorasib and docetaxel were essentially superimposed, maybe even the median overall survival for docetaxel was a little longer. So I think the efficacy is here. It’s a little disappointing compared to docetaxel, but I don’t think it changes how I think about using this drug today.

Beyond efficacy there is, of course, the consideration of toxicity, and I think we’re all very — we all understand and are comfortable with the adverse effects of docetaxel in the second-line setting. This is the data, though, comparing sotorasib and docetaxel in that second-line setting. And I think none of us need a reminder of the adverse effects of docetaxel, but you see significant rates of fatigue, alopecia, nausea, anemia, anorexia, and stomatitis for patients with KRAS G12C mutations who get docetaxel.

And by contrast, the major toxicities for sotorasib were some diarrhea, and I don’t mean to minimize it, you see a significant rate of Grade 3 diarrhea for patients with sotorasib, as well as ALT and AST elevation in about 10% of patients.

So I think if you look at this adverse event profile, and you talk about this with your patients, I think most of us are going to come to the conclusion that sotorasib’s a little bit easier to take, but I think it’s certainly a conversation to have with our patients.

So just to summarize for this randomized Phase III trial, sotorasib is superior to docetaxel based on response rate and primary endpoint of progression-free survival. The absence of an improvement in overall survival from the use of sotorasib compared with docetaxel suggests the benefits of this approach may be somewhat fleeting, but I think the toxicity of sotorasib is superior to docetaxel in this context.

The kind of new kid on the block for the treatment of KRAS-mutant non-small cell lung cancer is adagrasib. Adagrasib got its recent FDA approval based on the results of a Phase II trial, the data for which you see here. You see this waterfall plot, which is relatively similar to the waterfall plot we expect to see for patients with targeted therapy, a handful of patients over on the left side that aren’t benefitting, but the vast majority of patients having some tumor shrinkage.

One bit of data that was buried in the first New England Journal of Medicine article but has also been studied prospectively is the CNS efficacy of adagrasib for the treatment of patients with KRAS G12C-mutant non-small cell lung cancer. I think whenever we see these new targeted therapies kind of our first question is how do they work in the CNS, and kudos to the company developing adagrasib for actually prospectively studying that and beginning to get ahold of that question pretty early.

Again, this is data buried in the New England Journal of Medicine article. There are some caveats about this, and I wouldn’t take this too far as a retrospective evaluation of a prospective study. But you see here there is clearly some CNS activity with some patients having frank PRs or CRs, and it’s encouraging that there is CNS efficacy with adagrasib.

So to summarize, adagrasib is a newly approved drug targeting the treatment of patients with KRAS G12C-mutant non-small cell lung cancer. From single-arm study data there is a similar efficacy to that reported for sotorasib. We’re still awaiting the results of their randomized trial against docetaxel, and it’s certainly provocative data to suggest that CNS efficacy for adagrasib exists in patients with KRAS G12C-mutant non-small cell lung cancer. That doesn’t mean to say that it’s better than sotorasib, but we haven’t seen similar data for sotorasib.

DR LOVE: First let me just back up a little bit and ask you, I’m sure you have much more experience with docetaxel in the relapsed setting than these new drugs, do you see patients really benefit from docetaxel?

DR RIELY: It happens occasionally. I have to say I just met with a patient a couple days ago, when we reviewed her on-treatment scan on docetaxel, and there was a phenomenal response. But we have to admit that these are few and far between, and there is a very modest response rate, less than 10% response rate, and modest real clinical benefit for docetaxel in this context.

DR LOVE: Incidentally, do you usually add ramucirumab?

DR RIELY: About 50/50 on the addition of ramucirumab.

DR LOVE: So I’m just trying to figure out how the FDA works. We’re doing a program tonight on multiple myeloma, and there’s an antibody-drug conjugate, belantamab mafodotin, you may or may not have heard of, that just got pulled off the market. Single-agent response rate of 41%, but it didn’t beat — a pretty good drug, which I don’t even know why they put it against it because they usually use it afterwards, and it got — you can’t use it now.

So like I could go back to sotorasib and say there’s no survival benefit. I thought you need survival benefit to get a drug approved. And then adagrasib doesn’t have randomized data, right?

DR RIELY: That’s correct, yeah. No. I think trying to understand the FDA’s thinking is clearly an industry in and of itself. But I think these prospective single-arm Phase II studies, where we take a bunch of people with some oncogene and give them a drug, I think it makes sense for signal finding efforts. I think in rare — truly rare disease entities, things that are 1, 2, 3% of the patients, that’s probably the extent of the data we’re going to be able to have. But for something like KRAS G12C, where we have 10, 12% of patients with this abnormality I think it’s really —prospective randomized data are going to be the key for its value.

Now I think the — we can Monday morning quarterback the sotorasib trial for a long time to try to understand why it didn’t lead to an improvement in overall survival, and I think many of us will, but at the same time this adagrasib randomized data will come out over the course of the next year, as well, and I think that’ll be really helpful for our understanding of how we use these drugs.

DR LOVE: Yeah. I mean I don’t know what the right answer is. I’m just trying to understand what’s going on. I mean I certainly can understand why somebody would see a 1-arm trial and want to use a drug. I mean I don’t know that you need a randomized trial for something like osimertinib with that kind of response rate, but when you look at something like this it’s a lot more borderline, I think.

DR RIELY: What’s interesting is I don’t think any of us would have predicted that sotorasib would fail comparing overall survival. If you would have told me the front line — I’m sorry, the single-agent, single-arm trial data I’d say that’s pretty good. To me that’s clearly better than docetaxel. And then if you would have told me the response rate data from the randomized trial, which was a much better response rate, and you told me that an improved progression-free survival at 34%, I’d say that’s pretty good, and then they asked me to put down my bet on whether it would meet in overall survival, I would’ve put all my money on that overall survival being a positive result. The fact that it wasn’t, I really — I can’t quite figure that out.

DR LOVE: Yeah, but I mean the response rates are nothing like what you see in ALK and EGFR. They’re like half of that, right?

DR RIELY: Without a doubt. A quarter or even less.

DR LOVE: And even — yeah, and in the sotorasib study, yeah, you saw a hazard rate, but it was like 5 months versus 4 months or something. It was very — the absolute — so in terms of the accuracy of it I don’t know how accurate it is.

I’m mean basically I think, to me, it sounds like you’re looking at this as an equivalency trial. It has better toxicity, which makes a lot of sense except technically it’s not big enough, but it makes sense.

DR RIELY: Right. I think that’s definitely — when I look at the data that’s how I think about it now, is I think about it as a noninferiority with a toxicity benefit. But you’re absolutely right, that’s not the way it was designed. It was designed to show an improvement.

DR LOVE: Anyhow, we can talk more about it in the webinar. I just like try to figure it out as I go along here.

DR RIELY: I do think that the FDA — the approach the FDA is taking really — or actually the way I think about it is from the perspective of the people developing drugs. What they really rely on is the predictability of what the FDA is going to say and do, and in some ways the FDA has given us this model where we look at response rate in single-arm trials, and if it looks good, then we move it forward. Maybe the goalposts have shifted, though, with regard to what is good. And I think when we see drugs with response rates that are less than 40%, is that good enough? I think that’s a question we have to start asking.

But back to the predictability nature of the FDA, I think these companies are coming to the FDA with a product, and then they’re saying their design, and the FDA is saying okay, that sounds like a reasonable approach, and if you hit this mark then it’s likely we’re going to find an approval for you. And I think that predictability is important for the people developing the drugs, but it may tie their hands a little bit and make it so that we end up with some things that are not quite as exciting as we’d like them to be.

DR LOVE: So same question here as I was asking you about HER2 in terms of use of IOs in people. Are you generally using this first line? IO?

DR RIELY: Yeah. I think for patients with KRAS G12C-mutant lung cancer I think IO remains the standard first-line therapy, so a single agent if they have high PD-L1 or a combination with chemotherapy for the group of patients with any PD-L1 score, because we do see a significant higher rate of benefit for immune checkpoint inhibitors in this group of patients compared to any other oncogene-addicted disease. So this is a different kind of oncogene than EGFR or ALK. These patients are the classic patient with lung cancer. They have a much more frequent rate of smoking history in these much more genomically complex tumors, higher tumor mutation burden and that sort of thing.

DR LOVE: I don’t know if it’s just KRAS G12C or all of KRAS, but I think I remember seeing some data suggesting that globally you get, I think, higher responses with checkpoint inhibitors with KRAS-mutant disease, or am I thinking of something else?

DR RIELY: Yeah. I think it might be modestly higher. The data’s pretty modest in terms of how much data’s out there. We have some subset analyses from KEYNOTE-189 and that kind of thing, but it’s relatively modest data. And we’ve done some retrospective work here, as well, that suggests they are about the same.

DR LOVE: Great. Please continue.

DR RIELY: So we’ll move from KRAS to RET. RET fusions are of course one of the less common entities out there. This is some updated data for selpercatinib in RET fusion-positive non-small cell lung cancer. We all know that selpercatinib is approved for the treatment of patients with metastatic RET fusion-positive non-small cell lung cancer.

This is some updated data that we saw published this year. You see 2 categories of patients. On the left are those patients who were treatment naïve when they received selpercatinib, and on the right are those patients who had had prior platinum chemotherapy, so do we use this as a first-line drug or a second-line drug. I think what strikes you, and we’re beginning to see this as somewhat of a theme, is that there’s a higher response rate for first-line use of these drugs. So for selpercatinib response rate in the first line is 84% versus in the second line a 61% response rate.

But it doesn’t seem that — that response rate doesn’t really seem to translate to the more durable findings of response. And so if you look at median duration of response, in the treatment naïve patients the median duration of response is 20 months, and in the prior platinum therapy group it’s 29 months. And similarly, in terms of median progression-free survival, for the treatment naïve patient it’s 22 months. And for the patient who had had prior platinum the median progression-free survival is 25 months. So it suggests that, again, RET is something we’ve got to look for, selpercatinib is an effective agent that can be very effective in the first line, but it’s still got very good efficacy in the second line, both by response rate, as well as by progression-free and — progression-free survival and duration of response.

And just to summarize for that, so I think these are further data that support the efficacy of selpercatinib in RET fusion-positive non-small cell lung cancer. This higher response rate in the first-line setting really is tantalizing and tells me that I should definitely look for RET fusions in the up-front setting.

I’ve shown you efficacy data. I’ve shown you response rate, but one piece of data that we often overlook is quality-of-life data. The selpercatinib initial data had a quality-of-life component, and we’ve seen publication of that this year. Now I think most of us aren’t used to really digging into the details of quality-of-life data, but this is quality-of-life data. I would focus on the first bar. Each of these represents patients with lung cancer in the first bar. The second bar is treatment-naïve patients. The second bar is patients with prior lines of therapy, and then the third is more than 1 prior line of therapy.

I’ll just focus on the first bar. The group — that’s all patients with non-small cell lung cancer. And each group of 4 is at a different cycle. And so cycle 3, cycle 5, cycle 7, cycle 9, 11, 13. And you see the generally improved quality of life is present as soon as cycle 3 and gets just a little bit more over the course of these — really the almost 2 years that patients are on therapy. So really, to me, just additive data in addition to the efficacy data to say that these drugs are helpful in terms of quality of life for our patients.

Selpercatinib is not the only drug approved for RET fusion-positive non-small cell lung cancer. There’s also pralsetinib. We’ve seen some updated data for the ARROW trial that looked at the efficacy of pralsetinib in RET fusion-positive non-small cell lung cancer. And this is response data for first line and patients who’ve had prior platinum therapy. Again, you see for first line that we see a higher response rate, a 72% response rate, for patients who have RET fusion positive who get pralsetinib as a first-line agent. And in the second line or later, prior platinum group, the response rate is 59%. Again, most patients have response, but that response is not quite as deep in the group of patients who’ve had prior therapy.

These are the progression-free survival curves from the pralsetinib study. You see on the right the group of patients who had prior platinum-based therapy, and I think it’s a very impressive progression-free survival curve. On the left is a modestly complicated progression-free survival curve because they looked at the value of pralsetinib in the first line, but when they were seeing efficacy they — when they began the trial they said only patients who aren’t eligible for first-line platinum-based chemotherapy can get it first line, and then — and that’s what you see in red, that group of patients that were enrolled here. And the line in blue is that group of patients who were enrolled after it was allowed that patients — any patient could get first-line pralsetinib. And you see that that group of patients who were probably fitter, who could have also tolerated platinum-based chemotherapy, they did very well with pralsetinib as first line for their RET fusion-positive non-small cell lung cancer.

So just to summarize, pralsetinib for the treatment of patients with RET fusion-positive non-small cell lung cancer, we have continued evidence of efficacy for pralsetinib in patients in this context. The higher response rate is observed in the first-line therapy. It suggests up-front testing for patients with non-small cell lung cancer is critical to identify RET fusions and give those patients the best therapy up front.

DR LOVE: So I’ll ask the question that I seem to ask a lot nowadays, which is can you tell any difference between pralsetinib and selpercatinib in terms of efficacy or tolerability?

DR RIELY: There doesn’t seem to be a major difference in these drugs. I think if you — if I try to tease out the difference I think that I see a little bit greater duration of efficacy for selpercatinib versus pralsetinib. But these are cross-trial comparisons. I’m a little bit anxious to push that message, but in my experience I typically will be using selpercatinib in this context. That’s to say that the data for pralsetinib isn’t strong, but that’s been my comfort drug at this point.

DR LOVE: So that’s kind of the answer that I’ve been hearing for a while, but then your colleague, Dr Gainor, presented a patient to me who got selpercatinib, had a really kind of nasty chylous effusion, which I’ve never heard about. And then when I looked it up I see it’s not rare with selpercatinib. Have you ever seen it? And is that a concern?

DR RIELY: Yeah. I definitely have seen it. It’s certainly something to know about, something to pay attention to. I think that’s the big challenge for us, is that when we see an effusion on our scans we’re quite concerned that these effusions might represent a malignant pleural effusion and progression of disease, but this is a somewhat uncommon thing but definitely happens for patients who get selpercatinib. And identifying it and managing it are critical, even rarely seen other — beyond pleural effusion, other sites of chylous collections.

DR LOVE: The other thing is in terms of RET in general does it fit kind of a more classic definition of targetable phenotype? And specifically does that mean you’re not too excited about IOs? But also that you’re willing to treat brain mets.

DR RIELY: Yeah. This definitely meets those criteria. So it’s definitely lower TMB tumors, lower likelihood of benefit from checkpoint inhibitors, and very high efficacy in CNS. So this is definitely somebody — if I see a patient with a RET fusion-positive lung cancer who has brain metastases I’m perfectly content to give selpercatinib or pralsetinib as first-line therapy for that patient.

DR LOVE: All right. Please continue.

DR RIELY: We’ll move next to MET exon 14 splicing mutations. These mutations occur in about 3-4% of patients with non-small cell lung cancer. We currently have 2 drugs FDA approved in this context, capmatinib and tepotinib. Tepotinib received an accelerated approval based on single-arm data, and this year they presented the confirmatory results that had been requested by the FDA looking at tepotinib for an additional 161 patients.

And you see here on the left response rate and duration of disease outcome data; 55% response rate, median duration of response of 21 months, median progression-free survival of 14 months, and median overall survival of about 19 months. So this really tells us that tepotinib does have efficacy. It confirms that.

On the right you see how this current data compares with the older data. There’s certainly nothing worse about Cohort C and maybe slightly better than Cohort A, both in terms of duration of response and progression-free survival.

Once again, for tepotinib they looked at how do patients do based on line of therapy. And so within this larger cohort they had both first-line patients and second-line or later patients. And what you see here is, again, the response rate in the first-line setting is higher, with a response rate of 60%, versus 47% in the second line or later. There is, in this context though, a little bit longer median progression-free survival at 16 months versus 12 months, and a longer median overall survival of 21 months versus 19 months. So this again tells us that looking for MET exon 14 can be helpful so that we can give this treatment in the first line.

And so just to summarize for tepotinib, we have additional efficacy data for tepotinib in patients with MET exon 14-positive non-small cell lung cancer, and we once again see that the efficacy data support superiority of first-line use compared with later line in patients with MET exon 14-positive non-small cell lung cancer.

We’ve been talking about amivantamab for a few years. The first data we saw at ASCO, I’m going to say it was 3 years ago now, was for patients with EGFR classic mutations, exon 19 deletions, L858Rs, who had progressed after osimertinib. There was a respectable efficacy, but then we got further data and an approval for amivantamab in patients with EGFR exon 20 insertions.

But we sometimes forget what the molecule is, what is amivantamab? It’s a bispecific antibody, so 1 arm of the antibody binds to EGFR, the other arm of the antibody binds to MET. Kind of a unique concept. I think it was derived from the idea that EMT amplification’s important in resistance in EGFR. But nonetheless, we got this bispecific molecule that can hit EGFR and MET, and so kind of the obvious question is how does this work in patients with EMT exon 14-positive non-small cell lung cancer.

And so this year we got some of the first data around that. And this is the waterfall plot from the study of amivantamab in patients with non-small cell lung cancer with MET exon 14. And you see here that it’s kind of the classic waterfall with some response for most patients, a handful of patients on the left who didn’t get response, and then most patients having some tumor shrinkage.

Now one thing they’ve been helpful about is identifying what kind of patients they put in here. This is a relatively small trial with less than 100 patients, so you have to take these numbers with a grain of salt. But if you look at the overall patient population the response rate is 33%, which is certainly less than we see with drugs like capmatinib and tepotinib, but it has value.

But the kind of wrinkle is when you look in here the majority aren’t — almost the majority of patients had prior MET inhibitor. And in that group of patients, so typically that means they’ve had prior capmatinib, prior tepotinib, the response rate was much lower, 17%. By contrast, patients who had had — who had not had any prior therapy, either treatment naïve with a response rate of 57%, or just no prior MET inhibitor, response rate of 47%. So it really suggests to us that this drug can be effective with high response rates, 50% or thereabouts, but only before receiving any other EMT inhibitor.

So just to summarize, in addition to capmatinib and tepotinib the FDA-approved options for patients with MET exon 14 non-small cell lung cancer, amivantamab has some efficacy. We look forward to some additional data exploring its place in this group of patients. It’s got very modest efficacy after prior MET-directed therapy. And I’ll be honest, that’s not something I would have expected. I would have expected that since this has a very different mechanism of action that we would see that they would work — they wouldn’t have cross-reactive mechanisms of resistance, but here we are.

DR LOVE: So you talked about tepotinib, but you didn’t show any data for capmatinib. It’s basically interchangeable?

DR RIELY: They’re very similar drugs, very similar efficacy, very similar toxicity profiles. To my mind, I can’t really figure out a particular difference between the two.

DR LOVE: Please continue.

DR RIELY: One of the rare oncogenes is NRG1 fusions. NRG1 fusions, they probably occur in less than 1% of patients with lung cancer, but lung cancer is the place they’re most commonly found, though there’s a decent proportion of patients with pancreatic cancer who have them as well. And so NRG1 fusions are something that’s in the back of all of our minds when we do testing, and we occasionally find a patient with an NRG1 fusion.

Now NRG1 fusions lead to dependence on the HER3 pathway. We don’t talk about HER3 very much in treatment of patients with non-small cell lung cancer outside of EGFR-mutant disease. But there is a dependence on HER3 in these patients with NRG1 fusions, and this has been an idea that’s been in development for a few years.

And now we’re beginning to see efficacy data, and this is one of the first drugs with efficacy data. This is seribantumab. It’s an anti-HER3 antibody. And you see the waterfall plot here with patients who have NRG1 fusions who get treatment. This is 11 of the 12 patients here had lung cancer. The other 1-odd patient had pancreas cancer.

But you see here that there is a response rate with a confirmed investigator response rate of 36%. And again, most patients have some response to treatment.

So NRG1 fusions are a rare molecular entity. Typically you’ll find these by doing RNA seq type approaches but can sometimes be observed with conventional next-generation sequencing. These gene fusions lead to overactivation of the HER3 pathway, which provides an opportunity for use of drugs like seribantumab, which is an anti-HER3 antibody that’s efficacy is clear in these patients, though relatively small data thus far.

DR LOVE: So not so much for this program, but I wanted to know whether you would like to hear the only oncology joke that I’ve ever created in my life —

DR RIELY: I definitely want to hear it.

DR LOVE: — which is my suggestion for the trade name for seribantumab, or some other anti-HER3 antibody, what do you think about tras-threezumab?

DR RIELY: I like it. I love it. I think people would — I think it would be helpful on many fronts because people would really connect with the HER3 part, but it also connects with things they already know about. I like it.

DR LOVE: Let me tell you something. Somebody told me something the other day that I could not believe. Out of all the people I’ve talked to about EGFR am I correct in saying that HER1 is EGFR?

DR RIELY: Yup. It’s true. It’s true.

DR LOVE: I mean why didn’t they just call it HER1 then? That would have made a lot more sense.

DR RIELY: Right.

DR LOVE: Is that correct?

DR RIELY: Yup. It’s correct.

DR LOVE: It’s the same thing.

DR RIELY: It is. It is.

DR LOVE: Crazy. Okay. All right.

Anyhow, please continue.

DR RIELY: All right.

So the last bit of data I wanted to include about patients with KRAS G12C mutations is this data that was presented at a meeting that probably most of us didn’t — wasn’t on our radar, it’s ESMO IO. And this trial looked at adagrasib, a drug that we talked about previously that’s approved recently for the treatment of KRAS G12C-mutant non-small cell lung cancer, and they added pembrolizumab to adagrasib. Now we’ve seen data earlier this year of sotorasib plus pembrolizumab, and sotorasib plus pembrolizumab had a lot of toxicity challenges, and they weren’t able to give reasonable doses of sotorasib up front. And it led to a bunch of attempts to try to understand how to treat it better and how to combine them, but it was complicated, and there was no clear path forward. So I think we were all a little bit surprised when this data was presented at ESMO IO looking at adagrasib plus pembrolizumab.

I’m showing you first the toxicity data. So this is giving 400 mg BID of adagrasib, so a little bit lower than the full dose of adagrasib, but giving full-dose pembrolizumab. And you see here that the toxicity profile is pretty modest. There are none of the significant ALT/AST problems that we saw with sotorasib, though there still is about an 8 or 9% rate of Grade 3 elevation of AST and ALT, and there’s a little bit of nausea, a little bit of diarrhea, but all very manageable at this somewhat lower dose of adagrasib in combination with pembrolizumab. No Grade 5 adverse events and really encouraging in terms of safety.

They also reported efficacy data. Now this doesn’t look to me particularly different than adagrasib’s efficacy data. You do see that the vast majority of patients are having a response. Maybe the one area I’d focus on is over on the left, that group of patients who has the tumor growing. Very few patients over there with the tumor growing, so getting 2 approaches to the tumor, both with a small molecule inhibitor, as well as a checkpoint inhibitor really is a reasonable approach. They saw objective responses in 49% of patients across all PD-L1 levels, and in those patients with high PD-L1, 50% or greater, the response rate was 59%.

So just to summarize for adagrasib and pembrolizumab, I think it’s big news that adagrasib can be safely combined with pembrolizumab in KRAS G12C-mutant non-small cell lung cancer. And this is in contrast with what we saw with sotorasib, where we saw excessive toxicity when sotorasib is given with pembrolizumab.

Now there’s no clear efficacy advantage to giving this combination today but I think we’re going to see further studies that are looking at that.

And I’ll tell you, the place I put this in my head now is when I’m thinking about a second-line therapy for a patient with KRAS-mutant non-small cell lung cancer that patients is typically on first-line pembrolizumab. And when I go to think about my second-line drug I kind of lean towards adagrasib use now because I can be more comfortable that the toxicity of whatever residual pembro is there for that patient is not going to be such a big issue.

So I think this is really interesting data, and we’ll look forward to hearing more about whether we can combine these small molecule inhibitors with checkpoint inhibitors in the first-line setting.

DR LOVE: Any thought about biologically or pharmacologically why you would see — when you think about the way these drugs are built, so to speak, if it is true that you see toxicity with 1 and not the other, any idea why?

DR RIELY: I have to admit to being shocked that there was a difference because these molecules are not that dissimilar in terms of their structure, so I really did not expect that. So I don’t have a good answer, but I’m pleased that one of them can be combined so that we can explore this hypothesis.