Introduction: The Boards

DR LOVE: Good afternoon, everyone. I’m Neil Love from Research To Practice, and welcome back to Year in Review, as today we talk about nontargeted approaches to non-small cell lung cancer. We have a great faculty today, Dr Ben Levy from the Johns Hopkins Sidney Kimmel Cancer Center at Sibley Memorial in Washington, DC.

Today we’re here to talk about nontargeted approaches for lung cancer, and as always we will be discussing the use of nonapproved agents and regimens, so check out the package inserts for various products for more information.

So Dr Levy put together a great talk reviewing some of the key papers over the past year. Here are some of the papers that we will be discussing, as well as some of Dr Levy’s own papers.

Here’s where we’re heading. I’m going to just start out with a little introduction about my experience with the oncology boards, and then we’ll dive into some of these papers, and then end up, if we have time, hopefully talking about some of the research that Dr Levy is doing.

So Ben, people always came up to me and said, “Wow, your stuff really helps prepare for the boards.” And I figured I ought to go figure it out. I was kind of grandfathered in. I didn’t need to take the boards, but I took both the long one, and now I’m taking the every-3-month one. And I did that last weekend. You do it every 3 months, and — so I did the latest one this weekend.

And of course I’m not going to talk about any of the specific questions, but I just want to bring it up for a couple of reasons. One is I want to recommend that you take this every 3 months. This is a great exercise. They put a lot of work into this thing. Very, very well thought out. And you can look stuff up, you can use AI, whatever, but I recommend that you just sit down and take it because as soon as you give an answer, they go through with you the correct answer, why it’s correct, why the other answers aren’t correct, and in about an hour you really learn a lot.

So this kind of inspired me. So much great content nowadays. We spend most of our time talking about kind of what’s happened in the last few months, and of course the boards is a different story. You get into things that need to be vetted, it takes time. It usually runs a couple years behind. But again, this is really clinical scenarios and how you manage patients.

But there was several things that let’s just say I was inspired to bring up to you. They’re not all related to nontargeted, but I think they’re very interesting. So one thing is something — we’ve done so many cases of docs in practice since the pandemic. I think we’ve presented like 5,000 cases. A lot of lung cancer, but not too many SVC syndrome.

So I want to just give you an opportunity to chat a little bit, first of all about your own experience with SVC syndrome. Is it usually small cell? Do you see non-small cell? And also, how do you make the decision between whether or not to start systemic therapy right away or do radiation therapy? What kind of complications do you observe with SVC syndrome? And is immunotherapy — now that we have immunotherapy as part of first-line therapy, any complications of SVC syndrome that’ll make you nervous about using an IO? So Ben, what’s your experience nowadays with SVC syndrome? How often do you see it?

DR LEVY: Yeah. I don’t see it that often, and I’m not sure why. I think these are all good questions though. I’ve seen it — in the past when I have seen it, I’ve seen it both with non-small and small cell, and I think that distinction is important because management may be predicated on histology. Obviously, if you have a patient sitting there that’s got facial swelling and shortness of breath and a scan that suggests SVC syndrome, if they’re non-small cell, I’ve got to tell you, 9 times out of 10 I’m calling my radiation oncologist, and we are talking on the phone or chatting in my office about how to get this patient radiated quickly because historically speaking radiation has been the cornerstone for treatment for SVC syndrome when it's non-small cell. I think with small cell it’s a little bit more nuanced and different. We have to remember that small cell is an exceptionally chemosensitive disease, and that’s one scenario maybe when you see SVC syndrome, if it’s small cell, where it may not be inappropriate, so it could be appropriate, to start chemotherapy alone given that the response rates to chemotherapy with small cell are anywhere from 70 to 80%.

The third question about PD-L1, PD-L1 inhibitors, how to incorporate that in, I would say for a small cell patient I probably would hold off on the immunotherapy if I’m giving chemo given that the response rate is really what we’re looking at here, 70 to 80% with or without the immunotherapy.

How to incorporate immunotherapy in the non-small cell setting with SVC is a little bit more difficult because usually we are differing to the radiation oncologists first on that. Would I potentially consolidate and start chemo/IO after that? Certainly. But I think these are still some unanswered questions.

DR LOVE: Here’s another scenario. And again, we’ve been talking about this for years, but let’s just say I got inspired to ask you about it by doing the boards this weekend, and it’s the challenging situation of the patient with metastatic non-small cell, and you can tell me whether it matters whether it’s squamous or not, no actionable mutation, but the PD-1 is between 1 and 50%, particularly when you kind of get over 25%. And I was kind of wondering how you think that through. A patient says to you wow, do I really need to take chemotherapy? I’ve heard of this immunotherapy. So great, I’ve got a PD-1 score of whatever, 30%, et cetera, how you think this through. Are there people where you’ll actually use PD-1 alone? Do you think about how high it is even if it’s not 50%?

DR LEVY: Yeah.

DR LOVE: So just kind of curious. Would you ever start an IO and kind of see how a patient does?

DR LEVY: Yeah.

DR LOVE: And then if they don’t do too well add in the chemo? We’re going to talk later on about your work with cell-free DNA. Would that be something that would be helpful in a situation like this?

DR LEVY: Yeah. The world’s gotten complicated, much to the benefit of patients, given all the tools that we can leverage here. Speaking of the boards, the board answer here generally for a patient with a PD-L1 less than 50 is usually chemotherapy with immunotherapy or chemotherapy with dual checkpoint blockade. Those are probably the 2 most common regimens that are used. Now if they’re adenocarcinoma I would use a backbone of carboplatin/pemetrexed, and if they’re squamous I would use a chemo backbone of usually carboplatin and a taxane.

But you bring up a good point, and that is we need to remember that PD-L1 is a continuous variable, meaning that 40% is probably better than 30, and 30% is probably better than 20 when it comes to whether these patients are going to respond to immunotherapy. So it’s not an uncommon scenario for me, and this is medicine as an art and not the science of it, where I may have a patient who doesn’t want chemotherapy, whose PD-L1 is 35% or 40%, I don’t think it’s unreasonable in that setting if a patient says look, I get it, this is not where the data is, but I’d rather you spare me from the chemo. Let’s just start with immunotherapy and see how I do. We can always layer in the chemotherapy after 2 cycles if it’s not working. And again, medicine as an art not a science here.

The board answer is certainly chemo/IO or chemo/dual checkpoint blockade, I know we’ll talk about this, but yes, there are plenty of times in my clinic where I have these conversations. With a PD-L1 of 30 or 40%, older gentleman or older female, a patient with comorbidities who doesn’t want the chemotherapy, I don’t think it’s unreasonable to start that patient on single-agent IO, it’s not what I do for everybody, and then follow them carefully with scans and then make a decision whether you want to add the chemo later.

How to track patients with ctDNA. This is the holy grail. I think that when we think about CT scans it really is a Polaroid picture. The ctDNA analysis is getting down more to maybe a digital photo; kind of a lousy analogy there. But the bottom line is there may be a role in the future for us tracking ctDNA and using that to decide whether patients will get chemotherapy or not. Start them on pembrolizumab or an immunotherapy, follow their ctDNA, and if it doesn’t drop or it doesn’t clear maybe that’s an opportunity then to add chemotherapy.

DR LOVE: Okay. So here’s my other inspiration to bring up to you. And it’s really amazing how quickly things are moving, that if you’re a couple years behind kind of what you miss. I always notice this with the targeted therapy because things are really moving fast there. It was kind of interesting that with RET they didn’t even bring up lorlatinib, but obviously in the last couple of years that’s come onboard.

But the other thing that’s changed a lot in the last couple years has been first-line therapy of EGFR-mutant disease, and it just made me think a little bit about how the change in the approach, now that we have 3, before it was just osimertinib, now we have 3 options that are considered. And again, curious. For example, would you be much more likely to use chemotherapy in a patient with brain mets? Would you ever — I don’t know if you’re an amivantamab/lazertinib-type person, I’m not even sure what kind of CNS activity it has, but would that be a consideration? You guys really led the way in the whole idea of using systemic therapy for brain mets. I see it in other cancers that people pick up on this. But also it’s changed a little bit. So how do you think it through nowadays that first-line therapy’s different?

DR LEVY: Yeah. I think it through carefully. I think every patient is going to have to have an individualized treatment approach. We have 3 approved options now for front-line EGFR patients with sensitizing mutations in 19 or 21, EGFR 19 and 21 mutations: osimertinib, chemo plus osimertinib and amivantamab/lazertinib. I would say if a patient is on the younger end, brain mets we know portend worse outcome, I tend to favor adding in the chemotherapy to osimertinib with a shared decision-making with the patient to understand that there are more toxicities than just the osimertinib alone.

Right now I’m still learning how to use amivantamab/lazertinib. We have a lot of clinical trial experience with these — the combination of these 2 drugs. There are some toxicity issues that we’re still trying to wrestle with. That combination has exquisite activity in the brain, similar to chemo/osi.

So I would say that all 3 are options. They all have to be — all have to be integrated into this shared decision-making. No right answer here. I would say in my younger, fitter patients I would use chemo/osi. I think for my older patients maybe osimertinib alone. It’s going to be hard for me to think about an amivantamab/lazertinib patient right now; probably a very young, fit, motivated patient.

DR LOVE: Alright. So speaking of EGFR-mutated disease, something we’re really not going to talk about much, but yesterday we saw this, the FDA approval of Datopotamab deruxtecan, now first approval in lung cancer, EGFR-mutant disease. We’ll see what happens with nonsquamous moving forward. Of course in breast cancer we already have Dato approved in metastatic disease. And we’ll get into that a little bit later when we talk about ADCs, but to just kind of put that on the radar for us to make sure we talk about it.

Immune Checkpoint Inhibition for Localized Non-Small Cell Lung Cancer (NSCLC)

DR LOVE: Alright. So let’s get into some of the papers. I’m going to let you take it from here —

DR LEVY: Sure.

DR LOVE: — starting out with immunotherapy with localized disease.

DR LEVY: Yeah, a good place to start. I think we had some updates at ASCO this year on CheckMate 816. I think all of us are familiar with this study, patients randomized who had IB to IIIA to receive neoadjuvant chemo/IO versus neoadjuvant chemo then followed by surgery. So we know now that the complete pathological response to neoadjuvant chemo/IO is 25% versus only 2% in the chemo arm, but we had some nice updates at ASCO showing that there is a survival advantage with the strategy of chemo/IO followed by surgery versus chemo alone followed by surgery. I don’t think any of us are surprised, but it’s good to know that this all tracks and that further updates — the outcomes and the differences in outcomes hold, and this is a standard of care for patients.

Looking at OS in lung cancer-specific survival, not surprisingly here we see curves here that diverge early and stay diverted — and stay converged and creating — this was a new standard of care. This is what launched it all. This is from my colleague Patrick Forde. This came up at Hopkins. We were thrilled to be a part of this initial effort that was chapter 1 of a long story that’s being written now with neoadjuvant chemo/IO or perioperative IO for early-stage lung cancer.

The AEGEAN study was a very similar study. We’ve seen some updates recently from AACR and from World Lung. Again, patients with surgically resectable IIA to IIIB randomized to either chemo/durva in this case versus chemo alone. And on the back end, unlike 816, patients got a year of durvalumab, and in the control arm there was no immunotherapy either before or after surgery.

And we saw similar trends here. Improved major pathologic — excuse me — complete pathological response with chemo/IO versus chemo alone, improved major pathological responses with chemo/IO versus chemo alone. And we’ve seen ticks on event-free survival here, kind of similar to what we’re seeing with 816. We’re seeing curves that diverge early and stay diverged. We’re seeing improvements in event-free survival.

This really solidifies the role of immunotherapy not only before surgery but most likely after surgery, and we can certainly talk about that because it’s a little bit controversial. If patients get chemo/IO up front, and they have a complete pathological response, do they need the IO on the back end? And if so, if they don’t, who are the patients that need the IO on the back end? This is the updated analysis from AEGEAN that we have so far. Again, overall survival curves that look very similar to what we say from 816.

KEYNOTE-671, yet another perioperative strategy and approach, platinum doublet with pembrolizumab versus platinum doublet alone, again for surgically resectable IIA to IIIB. And we saw the same thing we saw with 816 and the same thing we saw with AEGEAN, very similar outcomes here. We see complete pathological responses around 20% versus only 4% in the chemo arm. And we’re seeing updated overall survivals. Once again, event-free survivals, overall survivals similar and consistent with what we’re seeing from 816, what we’re seeing from AEGEAN.

And then finally CheckMate 77T. This was very similar to 816, but it added the nivolumab on the back end of surgery for a year, which 816 did not do. So this is a true perioperative nivo trial, and once again resectable IIA to IIIB lung cancers randomized to chemo/nivo versus chemo, again surgery followed by nivo in the experimental arm, and in the placebo — excuse me — in the control arm no IO on the back end. Once again, we’re seeing these ticks in improvements in outcomes with this perioperative approach. This is not new to us anymore. We now have 4 or 5 studies that have shown this. So very, very exciting time for us to be in this field.

The question always comes up, now that we have perioperative neoadjuvant chemo/IO do we go neoadjuvant chemo/OI or do we go adjuvant. The world in lung cancer has always been adjuvant, and then all of a sudden now we have these perioperative approaches, and I was fortunate enough to be a part of this review article that really just highlighted the pros and cons of neoadjuvant versus adjuvant. And I would say the big thing, the big pro with neoadjuvant chemoimmunotherapy, number 1, is we know that patients are more likely to receive all 4 cycles of chemo neoadjuvantly versus adjuvantly. That is an advantage. But adding in the immunotherapy, of course, you’re eliminating micrometastatic disease before the surgery, you get an early assessment of tumor response. And again, I talked about the higher adherence. Whereas adjuvant chemoimmunotherapy you can go right to surgery, and some of these patients want to go right to surgery, don’t want to wait. I think right now based on the data we have we certainly have to pump the brakes given the benefit that we’ve seen in overall survival.

This is a picture just showing the neoadjuvant biology and really the importance of delivering immunotherapy prior to surgery, where you’re activating these tumor-specific T-cell responses that can rush into the tumor, and you’re eliminating the ability for these patients — for their tumors to metastasize to lymph nodes.

So this just kind of hammers home the strategy in a pictorial. And these are the practical implementations, the patient navigation. I’ll just say this, this is a busy slide, but this does take some navigation. The ability for a patient to see a surgeon and that surgeon then say hold on, we’re going to refer you back to the medical oncologist, and oh by the way we need the EGFR and ALK before we implement the chemotherapy with the immunotherapy. That takes a village. And I think it’s just so important that these cases are discussed in our tumor boards, that there’s multidisciplinary care, that we’re reaching out to the pathologists, we’re reaching out to the surgeon to get a sense — and even our nurses to understand how this is going to happen.

I think the big question still is if a patient doesn’t have a complete pathological response do we want to give immunotherapy on the back end? I think most of us agree we probably should do that.

Just shifting gears here a little bit, still focusing on immunotherapy but looking at immunotherapy more for all stages of lung cancer. This was a paper that we published out of Johns Hopkins looking at more than 11,000 lung cancer patients of all stages treated at Johns Hopkins. And we always think that autoimmune disease as these patients don’t do well, and we can’t give them immunotherapy. Well, one of the things that we did, again, looked at 11,000 patients, and roughly 4-5% of these patients had a rheumatological condition, and I think importantly we characterized what they were, and I hope that this is representative of what we’re seeing in the real world.

But we see just a smattering of different autoimmune diseases for these patients that were treated at Johns Hopkins, and I think we learned a couple of things. One is these patients did just as well if not better than the patients without autoimmune diseases. So I think we always want to not treat these patients aggressively and not give them immunotherapy, and I would make the argument that we need to consider giving immunotherapy depending on what the autoimmune disease is. But these patients also had earlier-stage diseases, which kind of runs counter to what we thought before.

And then another huge, Herculean effort that was recently done looking at immunotherapy for advanced-stage disease, now not early, stepping away from the perioperative prospective studies and looking retrospectively at patients with advanced-stage disease. As you can tell, there were lots of authors on this paper. It was a multi-institutional retrospective analysis. And the bottom line is this: what we found was that if patients had co-altering KEAP1 or STK11 mutations, these are mutations that will come up on every single next-generation sequencing report you do, there’s a suggestion that patients don’t do as well with the chemotherapy/immunotherapy as they would if they didn’t have a KEAP1 or STK11. So it seems like KEAP1 and STK11 creating this immunosuppressive environment that tells us that maybe chemo/IO is not enough.

The second clinical pearl from this study was we looked at the POSEIDON data, which we’ll talk about a little bit more in the program, but it basically showed that you can maybe get over the hump of the immunosuppression that this mutation carries by giving CTLA4 with the PD-1 drug. So maybe this is an opportunity when you see this, and I don’t think it’s for all patients, maybe it’s an opportunity to think that chemo/IO may not do the trick here. Maybe you need to pepper in or add in CTLA4 inhibition as well.

DR LOVE: Before you go on, maybe you can make a couple comments on the PACIFIC trial, and then I’ll bring up a couple questions we’re getting.

DR LEVY: Yeah. I mean, getting to the PACIFIC trial I just wanted to remind people, we talk a lot about neoadjuvant chemo/IO in early-stage disease and all these improved outcomes, but we need to remember that patients who aren’t surgical candidates, who are still curable, who get chemoradiation followed by a year of durvalumab, the 5-year survival rate is 43%. That’s bananas. That’s something we’ve never really seen in lung cancer. And oftentimes I think if a patient’s not a surgical candidate they’re still in good hands because we can give chemoradiation followed by a year of immunotherapy, so it’s a real exciting time.

DR LOVE: And I know there was this other paper, yeah, looking at real-world outcomes.

DR LEVY: Yeah.

DR LOVE: I always love to see this when it kind of supports what we’ve seen in the trials.

DR LEVY: It does, yeah. This was great. This was real-world outcomes through an expanded access program for durvalumab as consolidation post concurrent chemoradiation for unresectable lung cancer. And like you said, Neil, I mean, it’s great when the 5-year outcomes that we see in the real world are even maybe a little bit better than what they were in the trial. So that really reinforces and recapitulates this idea that patients who get through concurrent chemoradiation should receive durvalumab on the back end.

This is just showing you some of the outcomes based on subgroups from this real-world data. Not surprisingly the nonsquamous did a little bit better than the squams in this observational study. The PD-L1s greater than 1 did a little bit better than the less than 1s. The concurrent chemoradiation did a little bit better than sequential. Bottom line, though, is that these patients did really, really well with immunotherapy post concurrent chemoradiation.

DR LOVE: So there are a million questions that come up related to the papers that you just reviewed, but I’m sure people are going to go back and take a look at these slides and think about it.

I’m just kind of curious, though. Right now do you yourself look at KEAP1 and STK11?

DR LEVY: Yeah.

DR LOVE: And are you more likely to use CTLA4 yourself? I mean, I know it’s not in the guidelines, but what do you do?

DR LEVY: Well, I think it is an option. It’s certainly chemo with dual checkpoint blockade, with both nivo/ipi plus chemo a la CheckMate 9LA is an approved regimen for non-small cell lung cancer. I would say that this is a nuanced discussion. When we see KEAP1 and STK11 there’s no doubt based on the literature it is consistent that this creates an immunosuppressive environment, and patients that get chemo/IO probably don’t do as well as we think they would if they didn’t have these alterations.

I think the question here is when you add the CTLA4 yes, you may be able to improve outcomes, but that comes at a significant risk of more toxicity. And I think we have to ask the question, as I always ask the question, is the juice worth the squeeze. If I have some data to suggest that adding CTLA4 may improve outcomes is it worth the toxicity, and that’s where I think I have to have these discussions with patients. I think it’s really important to say adding a CTLA4 may help, you just need to be aware that the rash, the diarrhea, the LFT abnormalities may be more than chemo/IO.

Immunotherapy for Metastatic NSCLC

DR LOVE: Alright. Well, let’s move on now and talk about metastatic non-small cell but specifically related to immunotherapy, particularly updates on some of the key trials.

DR LEVY: So the 9LA study is kind of what we were just talking about. This was a study that we have updates on. This was a study looking at chemotherapy plus dual checkpoint blockade with both nivo and ipi versus chemo alone. Gosh, I would have loved if the control arm was chemo/IO, then we would be answering the question does CTLA4 added to chemo/IO do anything versus chemo/IO alone. That’s not the study. But the study did show in an updated analysis improved outcomes with dual checkpoint blockade with chemo versus chemo alone; median survival time 15.8 versus 11 with a hazard ratio of 0.73.

What’s interesting, I get this question all the time, outside of STK11 and outside of KEAP1, when would you use dual checkpoint blockade with chemo. And this is probably the most important slide to answer that question. The less than 1%, the PD-L1 less than 1%, there’s a really robust survival advantage. Now this was not powered to look at these subgroups, but that may be a group — you see a median overall survival time of 18 months, 17.2 months. That’s pretty competitive. That’s where I may think about using dual checkpoint blockade with chemo outside of the genetic alterations that we’re looking at. I also think squamous cell may be the one. I think we talk about this a lot in meetings, who is that patient that gets chemo/dual checkpoint blockade. I would say maybe an STK11/KEAP, squamous, and a PD-L1 less than 1 may be the way to go when you’re thinking about this.

The POSEIDON study was very similar to CheckMate 9LA, a little bit different with the arms, but asked the question, again, chemo/dual checkpoint blockade with durvalumab/tremelimumab versus platinum-based chemotherapy. There was also an arm that looked at chemo/durva, so just chemo/IO, so you do get a little hint here about what the addition of the CTLA4 did to chemo/IO regimen.

This was an update from January 2025. If you look on the left here, again, when you’re comparing both the chemo/dual checkpoint blockade to chemo, this is Figure A, you see a survival advantage, and when you compare chemo/IO you see a survival advantage when compared to chemo. When you begin to look at all 3 arms for both nonsquamous and squamous it does seem like this benefit was restricted only to the nonsquamous, which runs counter to the message we heard from CheckMate 9LA, which is that squam is the place to do the dual checkpoint blockade. It seemed like in this study, in POSEIDON, it didn’t make a difference. All 3 arms did about the same. So I think the jury’s still out on exactly the right patient to choose for dual checkpoint blockade with chemo.

This was a paper that I was happy to be a part of, and it’s really the holy grail. And the spoiler alert from the results of this paper is that we didn’t find what we were looking for but we still tried to find. And the bottom line is we’re always looking at what — who are these patients that live 3 years on IO and why are they living 3 years, whereas some patients who get IO live only 6 to 12 months or 1 to 2 years. So we tried to divide up 2 separate cohorts retrospectively and look at patients who’ve lived more than 3 years on IO and look at patients who haven’t done that well and lived only a year and tried to decide clinical and tumor and patient characteristics that would tell us differences. And these were the things that we tried to look at, the microbiome, the PD-L1 expression, the molecular profile, all of these things, the disease location, to try to tell us who are these patient who are living at 3 years.

And we didn’t find any real meaningful differences between these 2 groups, but nevertheless these are the efforts we need to find. We don’t know why some of these patients are living as long as they are on IO. Sure, a PD-L1 of 100% or a high TMB, no STK11/KEAP, sure, those patients are more likely to live longer. But these outliers, these — the tail of the curve, who are these patients? What makes up their tumor? What makes up their microbiome? What makes up the host tissue that we can discern the differences between that and a patient who doesn’t do well on immunotherapy.

DR LOVE: Yeah. I’ve always been fascinated by these exceptional responders. They’re always great cases to hear about. But it’s really amazing that you spent all this effort trying to figure out why, and maybe we still don’t have the answer.

One question about CTLA4. You’ve mentioned it a couple times. I’m curious in your clinical practice how using CTLA4, whether it’s ipi or tremelimumab, plays out clinically in terms of complications.

DR LEVY: Yeah.

DR LOVE: Do you see any difference between the 2 in terms of toxicity?

DR LEVY: Yeah.

DR LOVE: What’s the major issue with a patient with lung cancer?

DR LEVY: Yeah. Great question. I don’t see much meaningful differences in outcomes or toxicities with either ipilimumab or tremelimumab, so I think those are relatively similar. I do think there is a different toxicity profile for a patient who gets chemo/dual checkpoint blockade with either ipi/nivo or durva/treme versus a patient who just gets chemo/IO. And the things that I generally see in my clinic may be different than the paper toxicities that are on the trial, that we see on the result section of the trial.

I would say fatigue is one that we see more commonly, at least in my clinic, with dual checkpoint blockade versus immunotherapy alone or PD-1 or PD-L1 alone. I would say that rash and diarrhea tend to be more pronounced with dual checkpoint blockade. I would say LFT abnormalities are things that we need to look out for more. That may be more frequent. And I guess if you were to compare the 2 papers maybe they seem the same, but I can tell you in my experience it’s a different patient journey, and I think it’s so important that these patients are on a journey now. They’re not just getting one-and-done treatment, they’re getting sequenced strategies, and they’re living years, and so you’ve got to be so very mindful of the toxicities along the way and how to mitigate those.

DR LOVE: So Hassan in the chat room says, “What about the STRIDE regimen strategy in hepatocellular cancer where they give 1 dose of tremelimumab and that’s it?” Has that ever been looked at in lung cancer?

DR LEVY: Uh. It hasn’t, I don’t think. This has been brought up, can we get away with less? Do we need to integrate the CTLA4 on every cycle or can we drop it? Now in think in some of the studies they drop the ipilimumab after a while. I’d love to see just 1 or 2 doses to see if that primes the immune system enough so that you can kind of get out of the — get out of the CTLA4 inhibition. But these are the things that we still need to do, and we need to keep patients in mind. It has to be patient-centered care when we’re talking about some of these toxicities.

DR LOVE: So 1 more quick question. I really love that graphic where you looked at autoimmune diseases in patients with lung cancer. It really gives you a flavor of the frequency. I’m curious specifically particularly related to rheumatoid arthritis, are there situations where you really won’t use an IO or you really hold off on it?

DR LEVY: Yeah.

DR LOVE: What are your criteria to hold it?

DR LEVY: Great question. Great question. We are blessed here at Hopkins with a rheumatological division that has a significant interest in toxicities related to immunotherapy, and in fact the leader of our medical oncology group, who used to partner with them, Jarushka Naidoo, has led the way in characterizing the degree of side effects from immunotherapy and how to get these patients through immunotherapy.

I would say that in my mind it is so important that I speak to the rheumatologist for every patient. I do this for every patient. No matter what the rheumatological condition is I always speak to the rheumatologists who manage these patients, and I ask them, tell me what you think here. Do you think giving the immunotherapy’s going to flare this, and if it flares do you have the tools in your shed to cool it off? I mean, this is again medicine as an art not a science. I can’t tell you how important that is. I do this for endocrinologists, too. With thyroid issues I always call the endocrinologists. And I don’t put them on the spot, but I tell them tell me what you think here, because they’re looking at the patient through a different prism than I’m looking at them. And so every patient has a different cutoff for me, but I always consult with them when I’m doing this.

DR LOVE: Yeah. You mentioned Dr Naidoo. I love talking to her.

DR LEVY: She’s great.

DR LOVE: She’s like an encyclopedia when it comes to this.

DR LEVY: She’s amazing.

DR LOVE: It’s really great.

DR LEVY: Incredible.

Antibody-Drug Conjugates

DR LOVE: So let’s chat a little bit about ADCs. We talked about the approval that just came out yesterday with Dato. Here a couple of review articles that you were a part of that I just wanted the audience to know about. The slides are in the chat room if you want to pick that up and check out these papers. There’s a lot going on obviously throughout oncology with ADCs, but these are great papers on lung cancer and ADCs. Of course we have T-DXd for HER2-positive disease, but now we have your first TROP2 target now approved in lung cancer as of yesterday, Dato.

DR LEVY: Yup.

DR LOVE: Can you talk about the TROPION-Lung01 study? Of course the approval of EGFR was from the other TROPION study that focused on —

DR LEVY: Correct.

DR LOVE: — with mutations, mainly EGFR.

DR LEVY: Right.

DR LOVE: But this was a larger study. Can you comment on that?

DR LEVY: Yeah. I think we’re all disappointed in this study, although we’re still gleaning insights from it. And the bottom line is this was a very clean study looking at patients with advanced non-small cell lung cancer, both adeno and squamous, who had had disease progression on chemo/IO. If they had a target they could have also gotten targeted therapy followed by chemotherapy, so it was either second line or third line, randomized to Dato-DXd 6 mg/kg versus docetaxel, which kind of still remains king, incredibly; 20 years still in the second line.

So we thought that this would be a slam dunk homerun given some of the Phase I data we looked at with Dato-DXd, the preclinical work that’s been done and the potential that it targets selectively tumor cells and selectively delivers a payload. Very clean study. Randomized 1:1 in the second line. Unfortunately in the intention-to-treat analysis there was no benefit in progression — there was small benefit in progression-free survival but no benefit in overall survival. However, when you parsed out histology there was a benefit in overall survival for the nonsquamous patient population. And so despite that this did not move forward in the second line in an unenriched patient population, so I think we were a little disappointed in that.

What’s been interesting is this new biomarker, and I don’t have — I could spend 15 minutes on this biomarker alone. The bottom line is this biomarker is trying to measure cytoplasmic TROP2. The whole idea is the more cytoplasmic TROP2 you have, TROP2 protein in the cytoplasm, the more likely we feel that this payload is being internalized into the cytoplasm.

So when you look — it’s a tissue array. It’s a tissue analysis. It’s an artificial intelligence biomarker, a biomarker assisted by artificial intelligence, and the first in solid tumor oncology. So I don’t think we should sneeze at this. It’s still preliminary. The bottom line is in the TROPION-Lung01 study it did seem to tease out who did better with Dato-DXd versus if you were negative for the biomarker, but it didn’t seem to tease out those who got docetaxel. So this was truly predictive of outcome for Dato-DXd but not for docetaxel.

I was fortunate enough to lead the Phase IB study. I presented this as an oral at ASCO just a couple of weeks ago. The question is how does Dato-DXd do in the front line. Can it replace our beloved pemetrexed? Should we combine it with immunotherapy? Should we combine it with immunotherapy and chemotherapy? So this bottom line was looking at 2 separate sets of patients, patients who were treatment naïve who got Dato-DXd plus immunotherapy or — that was called the doublet, or who got the triple, which was Dato-DXd plus immunotherapy plus carboplatin, so again steering away in 1 arm from carboplatin and pemetrexed, and steering away in the other completely from pemetrexed.

And the doublet looked good. There was a median PFS of 11.2 months in the front line. These were 42 patients. This is a small dataset. The triplet didn’t look as good as the doublet, but that may have been due to the fact this is not a randomized study, and the triplet just happened to have more brain met patients and patients who were underdosed at a 4 mg/kg versus 6 mg/kg.

I think the interesting part, the most interesting part of this study was this: When we took that biomarker, that new biomarker, novel biomarker that’s an artificial intelligence-assisted biomarker, and we looked retrospectively at who did better in the front line it seemed the patients who were biomarker-positive, i.e. had more staining of their TROP2 in the cytoplasm of the cell, those patients did better than if you did not have biomarker positive on tissue. Very complicated biomarker.

I think we’ll learn more about it in future studies. There’s prospective analysis of this biomarker. The question here is, or maybe the point is can a novel biomarker help sort out who’s going to do well with this compound and who’s not going to do well. And if you’re negative, kind of like EGFR, maybe you don’t use it. And if you’re positive maybe you use it. Does this go first line? Does this go second line? Really tough to tell. Very interesting times right now.

DR LOVE: So yeah, the idea of an ADC plus IO in first-line metastatic is intriguing. We’ve seen that now with bladder cancer with enfortumab vedotin plus pembrolizumab —

DR LEVY: Yes.

DR LOVE: — beat up cisplatin combination. And now as of ASCO Sara Tolaney from Dana-Farber presented first-line sacituzumab, another anti-TROP ADC, plus pembro, probably going to become the new first-line therapy for triple-negative breast cancer. So we’ll see whether the same thing happens with lung cancer.

One question, though, getting back to Dato and this approval, I’m curious what your experience is with Dato in terms of toxicity. We’ve heard about mucositis. We’ve heard about ophthalmic issues, ocular issues. Of course everybody always brings up, anything with the second name of deruxtecan, ILD. I don’t think they’ve seen that much.

DR LEVY: Yeah.

DR LOVE: I know you’ve had some experience with Dato. I’m curious what your thoughts are, and for example how would you compare it to docetaxel or docetaxel/ramucirumab in terms of quality of life for the patient.

DR LEVY: Very different toxicities. I think the thing that you have to be watching out for and mindful of with Dato-DXd is the stomatitis. It is very common. In the study at least most of them are Grade 1 or 2, but there are Grade 3, and we need to do a better job mitigating that toxicity. I talked about the patient journey before and how important it is, when you have to answer that question is the juice worth the squeeze. And stomatitis is one of those challenges, where proactive steroid rinses, ice chips during the infusion may help mitigate that, and I think we’ll get there. We’re still learning, but a challenge. Docetaxel doesn’t have that.

I would say that the ocular issues are another thing we need to partner with our ophthalmologists. For patients who come in and say my eyes are really dry, that may be a keratitis or a corneal issue where we need to refer them to the ophthalmologist. ILD is not that — of a challenge for this drug. I’ve seen it a few times, but it has been reported. When I think about the journey of a patient on Dato-DXd, and I compare it to a journey of a patient on docetaxel, summing it all up, and I’m biased, I’ll just throw that out there because I’ve lead some of the studies with Dato-DXd, I think the journey’s a lot better with Dato-DXd because there’s — you can dose reduce, you can get the patients through the stomatitis.

I think we’re desperate to use something other than docetaxel or docetaxel/ramucirumab. My experience with that combination is okay. I’ve treated lots of patients over the past 10 years with doce or doce/ram. I think this is probably a little bit better. I think we know from the patient — from the TROPION-Lung01 that the nonsquamous patients there was an overall survival advantage with Dato-DXd. So if I had the drug, and it was approved, which it’s not by the way, but if I had it on my shelf I’d probably use it over docetaxel in the nonsquamous patient population.

Novel Bispecific Antibodies

DR LOVE: So I want to move on now and talk about really an area that we saw a lot of excitement about, I think it was about a year and a half ago, at the World Lung meeting, bispecific antibodies in lung cancer. So of course it’s — can you talk a little bit about what the bispecifics are and what are some of the agents now in development?

DR LEVY: Yeah. I think the bispecifics are just that. They tackle 2 different antigens on the cell surface of the cancer cell. They’re usually IgG based. We think about a monoclonal antibody with the 2 FAB regions, and they attach to the same antigen. A bispecific is going to have 2 different FAB regions, 1 is going to attach to 1 antigen, and 1 is going to attach to another, and consistently we’re seeing over and over again recently by doing this, by abrogating or inhibiting different signaling pathways within the cancer at the same time with this bispecific approach, we’re getting outcomes that I didn’t think — I don’t think we thought we would see.

Probably the golden child of bispecifics right now in lung cancer is ivonescimab. It’s a PD-1/VEGF bispecific. Who would have through VEGF would be back in play in lung cancer? Yes, we had bevacizumab a long time ago. Yes, we still have ramucirumab in the second line. But if you were to shown me this molecule 5 years ago and said look, it attacks VEGF and PD-1 I’d say forget it. Well, that’s why you play the game. That’s why you run the trial.

And this was a front-line trial comparing ivonescimab versus pembrolizumab, which is the standard, in advanced-stage patients that were treatment naïve with a PD-L1 greater than 1%. Now you could make the argument that pembrolizumab is probably not the ideal drug for PD-L1 1 to 49, but this was a global trial. It was actually exclusively done in China, where that is approved. And this is pretty compelling. I would have never thought this would be the case by inhibiting both VEGF and PD-1 you can get an improvement in progression-free survival like this, with a hazard ratio of 0.51. Again, pretty incredible to see these outcomes.

And when we look at toxicities, there’s not that much difference between — in my mind, when I look at this, between ivonescimab and pembrolizumab. Yes, it hits VEGF. Maybe there’s more proteinuria here, maybe a little bit more increase in LFTs, but overall the word on the street is that this compound is not that much more toxic than pembrolizumab.

We did have a press release in April showing that there’s an overall survival, hazard ratio of 0.77, but that the data, the OS is immature right now. So we’ll have to see how this all plays out, but ivonescimab may become a new standard. Now we need to run the trials in the US and across the globe; that was exclusively done in China. But we’ll see. Very, very exciting.

Two other drugs that are worth mentioning that are bispecifics, one is volrustomig, the other is rilvegostomig. I hope I got those names pronounced right.

DR LOVE: That was good.

DR LEVY: Very new. Very new in the field. But again, 1 tackling PD-1 and CTLA4, the other attacking PD-1 and TIGIT, and we’re seeing very encouraging activity with both of these compounds. This is data looking at volrustomig plus chemotherapy in the first line showing — you can see these waterfall plots here. Again, this is a PD-1/CTLA4 bispecific. Very encouraging data both in the nonsquamous and in the squamous. This is first-line treatment. It’s with chemotherapy. So we’re going to see future studies with this.

Toxicities are there. I would say rash, LFTs, thyroid. Similar things that we see with immunotherapy are there, and I can tell you that the bispecific approach with a CTLA4 and a PD-1 is much better tolerated than 2 separate drugs, PD-L1/CTLA4 put together, and that’s very, very encouraging.

These are the ongoing studies. You’re going to hear a lot more about this compound. This was featured at ASCO in terms of just kind of where to look for bispecifics. The other compound rilvegostomig, PD-1/TIGIT. We think TIGIT is out of vogue, but maybe it’s back in vogue. We’ll see. And there’s a whole program revolving around this drug. This was looking at this compound in patients that were immunotherapy naïve with a PD-L1 greater than 1%. We’re seeing response rates. You can see the waterfall plots that are here.

And again, you’re seeing toxicities, but TIGIT has never been this pathway when we abrogate it or inhibit it that we get a lot of toxicity. So overall very few Grade 3 and no Grade 4 or 5 treatment-related adverse events. There’s a tremendous excitement around these bispecifics, and I’m very encouraged. And I never thought that I would be so — I thought you would need a payload. I thought you would need an ADC, you would need to attach a chemotherapy moiety to these monoclonal antibodies, whether they attached that monoclonal antibody attached to the single antigen or as a bispecific.

It just may be that a bispecific in and of itself is enough without the payload. Look at amivantamab. That is a bispecific, and EGFR/MET-directed bispecific. Look at ivonescimab, a PD-1/VEGF bispecific. So there’s a lot of interest coming out on bispecifics. I’ve got to tell you that I’m starting to look beyond the payloads. Maybe we just get the right antigens targeted on the — the right ones and go from there because it seems like efficacy is not sacrificed, and it also seems like toxicity is a little bit better. So yes, a lot of excitement with ADCs, but certainly we’ll see with the bispecifics. A huge program revolving around this compound. You’re going to see a lot of studies read out very, very soon.

And then finally acasunlimab. This is yet another bispecific that has promising data. This is looking at a costimulatory — excuse me — a bispecific that inhibits PD-1 but also inhibits another pathway that actually stimulates the T cell, so it’s got both an inhibitory signaling and a proactive or trying to stimulate the immune system at the same time. And this compound’s been looked at in combination with immunotherapy and as a single agent in patients who are pretreated, who are refractory to chemo/IO. And we can see with the combination of acasunlimab plus pembro there’s median overall survival time of 17 months, and this is launching — and this is the safety here. And again, we’re seeing some of the same safety we see with other bispecifics, LFT abnormalities, fatigue, nausea, diarrhea, but most of these are Grade 1 or Grade 2 with very few Grade 3.

And this is the study, the ABBIL1TY study. It makes a lot of sense to compare this to docetaxel given the synergy that we saw in pretreated patients from the Phase I. So a lot of excitement here with the bispecifics. It’s going to be hard to keep up with them all. It’s also going to be hard to know where they all fit in. But very, very exciting.

DR LOVE: Yeah, your thought about do we need a chemotherapy payload is really a fascinating one. Obviously we could avoid a lot of toxicity if we didn’t have to use that type of payload. And I remember somebody in the past was talking to me about maybe in the future we would have ADCs with an immune payload.

DR LEVY: Yes.

DR LOVE: But I guess in a way that’s kind of like what you’re talking about, which is a bispecific.

DR LEVY: That’s right. I mean that’s the question, is the bispecific in and of itself enough to elicit an immune response, especially when you’ve got signaling that you know is relevant with harnessing the adaptive arm of the immune system. You’ve got the CTLA4 and the PD-L1 drug, the antigens being locked in by the same molecule, or VEGF and PD-1 or TIGIT and PD-1. Maybe you achieve that with just 1 molecule without the payload, where you’re able to elicit the response rates. Maybe it makes it easier to combine with chemotherapy versus an ADC. Maybe it makes it easier to combine with targeted therapies. This is where we’ve learned from amivantamab it can be combined with lazertinib, and maybe that’s the way we have to go rather than using a payload or an ADC. As much as I love ADCs and participate in the science I’m starting to rethink payload versus no payload with some of the efficacy results we’re seeing.

DR LOVE: So yeah, when you mentioned who would have thought that VEGF was going to still be relevant I immediately thought of John Heymach —

DR LEVY: That’s right.

DR LOVE: — who for many years has really been pushing that pathway, and I’m curious to know what he thinks about these new bispecifics.

Journal Club with Dr Levy

DR LOVE: So the last thing I want to ask you about is some of your work, particularly this paper in Nature Medicine. I feel like in almost every solid tumor program we do, and some heme programs, in the chat room people always bring up cell-free DNA. A couple people brought that up here already. And you’ve done some really interesting work in that. Can you talk a little bit about this paper and in general what you’re looking at?

DR LEVY: Yeah. Well, I think that this is a nice kind of round out to the hour and the huge effort on the part of mostly our translational scientists, and Elsa Anagnostou, who’s leading the ctDNA efforts across the globe. And the bottom line that we’re trying to look at here is — and ask the question is, is it reliable to track ctDNA in patients who are getting immunotherapy, and can we use that information to make treatment decisions. And do we need to wait for a scan? Can we use what we’re seeing on a ctDNA longitudinal analysis?

Now this is not MRD, right? That’s another way to look at it. That’s postsurgery, and you’re cured, and then you track to see if something’s coming up and then maybe make a decision. This is in the advanced-stage setting, where you’re taking patients, and in this study we took 50 patients, and we just basically treated them with pembrolizumab. These patients all had a PD-L1 greater than 1%, got pembrolizumab, and then tracked their ctDNA over time. And this was just trying to get an understanding of what ctDNA does for patients on immunotherapy.

And the bottom line for that’s really interesting, a couple things. One, this is not a bespoke ctDNA. This is just looking at the variants of interest for lung cancer, so we’re not using tissue informed. But we were able to really characterize 4 different patterns, basically what are the groups doing. And so there was certainly a group of patients who had complete elimination of their ctDNA after 2 cycles of treatment, which is pretty remarkable. Then there was a group that had complete elimination after 3 cycles. Then there was a group who basically had an 85% reduction. And then finally there was a group that never dropped. Their ctDNA just never dropped when you gave the pembrolizumab.

And what’s really interesting here, and we report this out in different figures and in different iterations, and I’ll just kind of give you the summary and then go through the slides briefly, but the bottom line was the response that patients had to ctDNA correlated with how they responded radiographically. So if you had a patient who had a drastic response or dramatic response on the scan there was a good chance that they had complete elimination of their ctDNA before that scan was even done.

And so the question here is, can you use this information before a scan is done. And this is going to be rolled out into another study that we’re now doing, which is essentially taking patients who are getting pembrolizumab, and if they don’t clear their ctDNA after cycle 2 or cycle 3 we add in chemotherapy. Why wait for the scan? If we know you don’t clear why not do something early on? And maybe this is how we’re going to use this in real time.

And these are just multiple different graphics showing you that the drop in ctDNA clearance, and even the depth of response correlates with RECIST radiographic response. So the deeper the drop in ctDNA the more likely the tumor was to shrink and shrink more dramatically. So there’s a nice correlation here, and we really look forward to basically having this played out in real time in Phase III.

My hope is, Neil, that there comes a day very soon, maybe 2 years down the road, where if a patient’s on immunotherapy, or even targeted therapy, at 3 weeks or 6 weeks I can do a simple blood test and say, “Hey, guess what? You cleared your ctDNA. You’re good. We don’t need to do anything. Let’s keep you on the drug. Maybe we stretch out your intervals of scans to 6 months rather than 3 months.” Those patients who don’t drop, however, that’s where we say, “Look, maybe we need more frequent monitoring. Maybe we need to add in chemotherapy right away.” So that’s the take home from this paper. I feel passionate about the use of this. I think it’s going to change the game for us in the next few years.

DR LOVE: Maybe it’s going to be on the boards in 10 years. Who knows? It could be standard of care.

DR LEVY: Yeah.

DR LOVE: But anyhow, I’ve got to squeeze in 1 quick case from the chat room because Emily put this in twice.

DR LEVY: Sure.

DR LOVE: So I have a feeling this is an actual patient she’s trying to make a decision on.

DR LEVY: Okay.

DR LOVE: I’m going to run it by you real quick.

DR LEVY: Do it.

DR LOVE: Heading back to the durvalumab consolidation.

DR LEVY: Yup.

DR LOVE: 74-year-old woman, recurrent adenocarcinoma of the lung, Stage III unresectable, with KRAS G12C, got chemoradiation therapy and is “not a candidate for an IO” because she had pneumonitis, went on pembro as adjuvant therapy. Interesting.

DR LEVY: Right.

DR LOVE: We were talking about adjuvant IOs.

DR LEVY: Yes.

DR LOVE: She got pneumonitis.

DR LEVY: Yeah.

DR LOVE: So she doesn’t want to give her durva, so she wants to know what about KRAS G12C inhibitor as consolidation?

DR LEVY: Great question.

DR LOVE: Or just the observe?

DR LEVY: I will tell you; we’ve had this several times. Patients who’ve received prior IO in the adjuvant setting who have gotten pneumonitis, don’t want to pull the lever on consolidation durva after chemorads because they’ve gotten radiation to the lung, too, I’d probably use observe in that setting. I wouldn’t give a KRAS-directed G12C. Those drugs, right now there just not as good as we want them to be. We’re so close with the KRAS G12C-directed therapies. But in that setting, if it’s truly a contraindication to give immunotherapy I’d probably just observe and watch closely.

DR LOVE: So 1 final quick question. I think pretty much people have moved over with EGFR and ALK to using post chemoradiation targeted therapy there. What other targeted therapy would you use as consolidation rather than durva, ROS1, BRAF, et cetera?

DR LEVY: Right. What a slippery slope. This is a tough one. I would say that every physician here may have a different answer. I would say for EGFR and ALK post concurrent chemoradiation osimertinib and alectinib would probably be it. Keep in mind, alectinib’s not approved post concurrent chemoradiation, but we use it based on the ALINA data. I would say for a ROS1 or a RET fusion I’d probably — again, not approved, not FDA approved, I’d probably considering using a targeted therapy in that setting.

I think outside of that the jury’s still out. The jury’s out on everything other than EGFR, by the way, but I would try to get at least a ROS-directed therapy or a RET-directed therapy post concurrent chemoradiation. When you start talking about BRAF V600E and MET exon 14 skipping and HER2, yikes, it gets a little bit more challenging to make those decisions.

DR LOVE: So Ben, thank you so much. It’s always a pleasure to talk to you. I always view Hopkins, my alma mater as you know, I think, as checkpoint central for the world, and I think some of the work that you’re doing now just reinforces that fact there’s so much incredible research going on a Johns Hopkins.

Audience, thank you for attending. Be safe, stay well and have a great night.

Thanks so much, Ben. Great job.

DR LEVY: Thanks, Neil. Thanks.