Evolution of targeted therapy for HER2-altered non-small cell lung cancer (NSCLC)

DR LOVE: Welcome to Oncology Today: The Management of HER2-Altered Non-Small Cell Lung Cancer. This is medical oncologist Dr Neil Love. For this program, I met with Dr Bob Li from the Memorial Sloan Kettering Cancer Center in New York City. To begin our conversation, I asked Dr Li to review the evolution of targeted treatment for patients with non-small cell lung cancer.

DR LI: It really started, at least in lung cancer, from the EGFR story that almost 2 decades ago. And the Tom Lynch paper was really a major landmark publication, that sea change. It took another 5 years or so, before it got translated into saving lives, because back then the EGFR inhibitors were developed for all-comers. There was a saying, I think I picked it up from your podcast, everybody get dose of erlotinib before they’d go to hospice. And I think Corey Langer, you did an interview with him, and he had — I had a driving to work and I heard this. So that was how it discovered, the drugs were discovered before the targets were really found.

And that general inhibition just doesn’t work because it’s not the right binding into the molecule.

So the EGFR mutation discovery, and at the time Tom Lynch, along with a group at Dana-Farber. So, Tom Lynch was at Mass General; Dana-Farber, they were sort of a close second. They had a publication in Science about 2 weeks later. And then MSK was also working on the same thing, back then, with Harold Varmus and William Powell and Vince Miller and others. And it was really 3 groups simultaneously, showing the same thing. And that was around 2004. But it took until 2009, when the IPASS trial as published in the New England by Tony Mok, Yi-long Wu and others, and that was a Pan-Asian study in EGFR-mutant lung cancers, were seen more common in Asian women who never smoked. And they had an advantage in terms of enrolling these patients. And that led to a complete change of mindset in thoracic oncology.

So the clinical trials were the ones that translated that discovery into clinical practice and saving lives. And of course, that then led to subsequent generations of EGFR inhibitors as well as ALK, multiple generations of ALK inhibitors and ROS inhibitors.

They all share the same of mechanism of action. It’s about this lock and key approach. If you find a mutation in the oncogene, and that drives the oncoprotein into uncontrolled cell proliferation and metastasis, and you see that there is a cleft or a pocket in the oncoprotein, and that you fit a drug into it and to lock it up, so to speak. And then it just suddenly stops that signal and then the cell just stops dividing and it dies.

So that’s the, sort of concept behind original thought process in the targeted therapy world in lung cancer.

However, things are now moving beyond that. So when we get to HER2, people have tried that lock and key approach, but it just hasn’t worked out. It’s actually been studied for about 20 years, ever since the trastuzumab revolution in breast cancer in the year 2000 or so. And for 20 years we’ve not been able to develop a HER2-targeted therapy in lung cancer because, firstly, just replicating the trastuzumab trial from breast to lung didn’t work out because they were different. And secondly, the incidence of HER2 expression and amplification, for example, is vastly different; the mechanisms are different.

And the other aspect is, the lock and key approach just doesn’t work too well for HER2 mutations, at least the early generations of drugs didn’t do a perfect bind. These were insertions in exon 20. And there’s no pocket. It’s a different shape. And you don’t have a nice key that can just lock it up.

So that’s why the tyrosine kinase inhibitor trials have largely failed whether it was neratinib or afatinib or dacomitinib. And we published those trials and they were sort of showing some modest responses in the 10- to 20% range, but nothing outstanding, nothing that could beat chemotherapy. And, therefore, it’s been disappointing.

But to go beyond that, science of HER2 activation in lung cancer is seen as different. A lot them were driven by mutations that are hyperactivated that leads to internalization of the tyrosine kinase and the receptor, the HER2 receptor. And in a way, without having too much HER2 protein on the cell surface, they kind of just go into the – they hyperactive, they internalize into the cell and then it gets recycled.

So if you attach an antibody drug conjugate onto it, it literally just sucks the drug into the cell.

You can see that ADCs can get into the cell through a variety of mechanism. The traditional understanding is the one on the left-hand panel. You have lots of HER2 protein on the cell surface. You just need a lot more for the drug to bind and, therefore, get in. That’s the traditional dogma. And that’s why a lot of the very prominent scientists and physicians in this field have warned me, you’re not going to – you’re going to get a negative trial with the HER2-mutant story because there are not that many receptors on the cell surface. But my curiosity sort of prevailed and we still did the trial anyway. And it was positive.

So that then led to the second mechanism of action where there’s not that many receptors. There are always going to be some, but not amplified or overexpressed. And what they do is merely just sucking the drug actively into the same, at a much more internalized rate than the wild-type receptors.

So we call that internalization through ubiquitination. But it’s really literally just sucking the drug into the cell. And these mutants are very good at just internalizing the drug. And if you have a precision guided missile with a warhead, or the powerful chemotherapy that is incompatible with life if you just give it naked, but if you pack it into the antibody and the cancer cells suck that drug into the cell and then you release the bomb from within and kills the cancer cell, that kind of precision missile attack actually worked. And we initially found that in case reports. There were case series from Europe, Julien Mazières and others from France, had published. And those were anecdotal. The lab evidence was lagging.

So we decided, there was some preclinical hypothesis that we came up about this internalization theory but it wasn’t proven at the time.

So we did concurrent lab and clinical research experiments.

We then took the clinical practice back to the labs, so it’s research to practice and back to the lab to research. And we can then label these antibody drug conjugates with rhodamine dye, and we call it pHrodo, because it’s PH-driven. And then they sort of glow in the dark and then you can see the rate of internalization of these ADCs into the tumors.

Here, we were able to show the one that I labeled, the ADCs going into the cell and real-time.

So if you go to the next slide, we can then see the rate of internalization of these ADCs into the cell at different rates. So we call it sort of the trafficking index. That’s how fast these drugs can get into the cell. So some of the cells literally pulled the drug into the cell and they traffic very quickly. And others are very slow at doing that.

So that was sort of the mechanism of response that we looked at. And then we then took this to the clinic again and that led to the DESTINY-Lung01 and the registrational trial approval just a few months ago.

And there was a lot of skepticism at the time, this was 2014, that many HER2 experts who really spent their lifetime doing tyrosine kinase inhibitor cell signaling and lock and key approach or people in breast cancer who were really looking at the quantity of HER2 and its driven by HER2 amplification and overexpression rather than mutation. The concepts are very different and they just thought that the ADC approach is never going to work in HER2-mutants in lung cancer.

So it came as a surprise when the T-DM1 trial was positive. And then that led to a lot of interest from biopharma industry to invest in this rather obscure field when it first started. I mean these are still 2% of lung cancer. You look at ROS1 and ALK, it’s not that many more in terms of percentage incidence. We’re talking thousands and thousands patients every single year just in the United States alone, and worldwide it’s a lot more patients.

So then it led to some interest in this field. And then, strengthening from the T-DM1 pilot study which was published in the Journal of Clinical Oncology back in 2018, we did a series of translational experiments that show that this was not a fluke — this was a real effect that was replicated in cell lines, in mouse models. And some of the slides, I’ve put some pretty pictures there just for reference. And we then were able to guide the registrational trial of trastuzumab deruxtecan, which was first developed for HER2-amplified and then HER2-low breast cancer, but it's all largely due to protein expression, not mutations.

But because of this internalization concept that we had published and shown subsequently, we doubled down on that in lung cancer. And that led to the registrational approval of the first targeted therapy for HER2. But it’s a different mechanism. It’s a smart bomb and precision missile, compared to a lock and key approach.

Incidence and management of HER2 amplifications and mutations in NSCLC

DR LOVE: And it was interesting, the last ASCO meeting when we had the HER2-low data presented and Dr LaRusso did the discussion, and she brought up the issue of HER2-mutant disease. And when I would bring that up to the breast cancer docs, they were like, well, we hardly ever see it. And it’s kind of a not a major issue for them. Even in GI cancer where, again, T-DXd and HER2-positivity is a big issue, colon, upper GI. I think it’s mainly overexpression there, also.

So, as you say, it’s a really different concept. First of all, can you just kind of maybe outline a little bit about, I guess the epidemiology. Like, what fraction of patients with non-small cell lung cancer have HER2 mutations as opposed to HER2-overexpression? For every patient with HER2-overexpression, how many are there with HER2 mutation? Is there like 10 times as many? It’s a lot more I guess.

DR LI: Yeah. Yeah, you’re right. The epidemiology is different by disease. So, compared to breast cancer, where 20% are HER2-amplified, which then produces the expression. The generic aberration is amplification, and that’s just increased gene copy number in breast and gastric. And I tell my patients, all my fellows, that this is like going to the old-fashioned photocopy — if they’re a bit older I can explain the photocopy machine — and you press on the button and you forget to let go. And then you’re chatting away and you realize you copied 50 pages and it was a total waste of money and paper and the environment. That is amplification. There’s no difference in the genetic coding of the sequence. And basically, it’s the same thing but copied again and again and again. And we call it copy number gain. And when it gets to a certain threshold, we call it amplification.

Mutation is different. It’s not just a quantity of the protein that’s been expressed or copied, the sequence is different. They’ve done an error. It’s like they wrote the wrong thing. And the DNA sequence is wrong. And that could be wrong because you’ve switched a letter for another, so that’s sequence nucleotide variation, or you can right a wrong in the sense that you’ve put in, you’ve inserted a list, like 12 extra letters in between the sentence. So that’s called an insertion. Or you can just delete some stuff; you forgot you deleted it, and that’s a deletion.

So these are so-called mutations. And that is different because the wording is different. The content has changed. But that change in content has led inadvertently to activation of the gene in uncontrolled fashion.

So that’s a slightly different concept. So you’re not going too many copies. You’re not going to get too much protein on the cell surface. You’re just going to get a different HER2 that functions crazy, it just grows and divides and uncontrolled. It just functions wrong. But the number of copies of HER2 is still the same.

So, in that sense, in lung cancer, it’s only about 2% — 2% are mutant, in terms of total non-small cell lung cancer. In lung adenocarcinoma, probably 3%.

And you do have a 1- to 2% also incidence of HER2 amplification in lung cancer. So that does exist, but nowhere near as frequent as the 20% in breast and gastric cancers.

So you have amplification in lung cancer and you have mutation in lung cancer. And according to our translational studies at Memorial Sloan Kettering and elsewhere, those 2 are largely separate targets; they represent different patient populations. There may be a small overlap in the order of 5- to 10%, but they’re largely separate.

The mutation cohort of patients tend to be younger people, often women, and often never-smokers. And those, it’s a pretty — there’s some analogy to EGFR-mutation in terms of epidemiology. But it’s not predominant Asian. There is no racioethnic predominance for HER2 — I’ve seen HER2 mutations across patients of all racioethnic backgrounds. And it’s not related to smoking. It’s something in the environment that may have triggered it or just a combination of multiple factors with a lot of bad luck. And I’ve seen patients — there’s no sort of single exposure that I can pinpoint. I’ve seen patients who have — one of my earliest Australian patients came from a farm with very clean air and was no industrial pollution. But he had HER2-mutant lung cancer and was pretty young as well.

So it’s a different category of lung cancer. And that’s about 2%. But when you talk about amplification, it’s really a miscellaneous group of diseases in lung cancers. You do have the heavy smokers, and some of them have HER2-amplification. Some of them are squamous. But, also, up to about a half of them actually have EGFR-co-mutation. So you then have a HER2 driver and an EGFR driver, and often the HER2, initially thought to be a resistance mechanism to EGFR inhibitors. But subsequent research has shown that it’s not always a resistance. Some of them had HER2 amplification at the outset.

So you really have this molecular heterogeneity in HER2 in terms of its oncogenesis in lung cancer.

So I’d categorize them as mutation as 1 group. Amplification as another. And it’s really the amplification that drives the high level of protein expression. And that’s how I see them in terms of therapeutics, but also in terms of drug development and clinical trials.

DR LOVE: Well, also there’s the issue of the phenotype of the patient and some sort of corollary issues. So, just to pick out 1 is sensitivity to immune checkpoint inhibitors. Actually, I think I saw a paper that you did — I think it was HER2-mutant, showing no responses. And that’s kind of what we hear with ALK and EGFR. Is that the viewpoint? Or is that the way you see HER2-mutant — not necessarily very responsive to IOs? And then, what about the HER2-amplified? Are they sensitive to IOs?

DR LI: Our early analysis of this group showed that the IO responses tend to be low compared to the general non-small cell lung cancer population. And it corroborated with the European experience with Julien Mazières Annals of Oncology, that series had shown a 7% response rate. So it’s not 0. Some of them do respond. But it’s actually lower than what you expect with the general population.

So single-agent immune checkpoint inhibitors, I tend to reserve later-line rather than upfront for HER2-mutants.

There’s a German series in the Journal of Thoracic Oncology, that showed that if you do combination chemoimmunotherapy, they may do okay. But it’s hard to tease out whether it’s the chemo that did it or is it the immuno that did it.

So we’re in the process of running a randomized Phase III trial of comparing first-line chemoimmunotherapy versus trastuzumab deruxtecan, T-DXd, in the first-line setting. And because it’s such a rare mutation, we really had to get the world together to do this and to get meaningful timeframe in terms of reading out results. So, hopefully, through international collaboration, we can get the results in the next couple of years to give us the answer for optimal first-line therapy.

DR LOVE: It’ll be interesting to see what happens there in terms of PD-1 levels.

Role of ErbB/HER protein family biology in lung cancer 

DR LOVE: Maybe you can take a shot at — I’ve asked a lot of people to explain to me, and I’m going to let you take your shot at it, sort of the biology of not just HER2, but also EGFR. Like, recently I became aware — am I right in saying that HER1 is actually EGFR?

DR LI: Yeah! That’s right.

DR LOVE: So I didn’t know that. but anyhow, it just reflects my lack of knowledge about that whole axis. And HER3 and the fact that there’s not a TKI. But anyhow, can you just maybe explain it the way you explain it to fellows, your vision of that whole axis?

DR LI: Yes. Thank you. It’s a good point you just made. Academics work in too many silos and they don’t necessarily talk to one another. And then you have different lingo coming out of their publications.

So EGFR is HER1, and that’s the neighbor of HER2. And HER2 is also called many things: HER2/neu, ERBB2. And these are just because of different academic groups doing their own thing and calling it their own things. And then, in the scientific literature, it gets very confusing. And that’s just an effect of different academic silos and not collaborating well enough. But hopefully, Research To Practice will inform and help break those silos so that we are talking about the same thing. And that way we can advance this field of science in much expeditious fashion rather than keeping it to your chest, in our own little world. That’s what academics have an inclination towards.

In the big picture, these are all so-called human epidermal growth factor family of receptors. So those are HER1, HER2. There’s HER3, HER4. And they all act — they are all proteins on the cell surface. They serve as signaling growth factors. So the cells have to talk to each other with signals. and once you get a signal, the HER family of receptors are responsible for sending those signals into the nucleus of the cell for DNA replication.

So they sort of control — and they have to do their job in terms of growing: when you need to grow and stopping the growth when you don’t need to grow so we don’t grow forever.

So that’s the normal part of human biology. All our cells in our body have HER receptors on the cell surface. Otherwise, we wouldn’t be able to survive.

But when the genetic codes of HER2 or HER1 or EGFR make a mistake, such as a mutation, an error in the coding, then those HER receptors, whether it’s HER1 or HER2, can misbehave or just grow uncontrolled. And that’s when it starts to send those signals down to the nucleus and telling that specific cancer cell to grow and divide. Meanwhile, the rest of the normal cells are still having normal HER2 or HER1. But the mutant cells in the cancer, they want to take control. So they grow and divide.

So the HER family is really integral and it’s been a very important target for cancer therapeutics. EGFR is certainly the poster child for lung cancer precision medicine, HER1. And HER2 is the poster child for breast cancer precision medicine with all the revolutions since the days of trastuzumab, but certainly many iterations of trastuzumab-related therapy.

And these have transformed lung cancer and breast cancer survival in really amazing ways.

So that’s the paradigm. And it’s going back 20 years or so since the original discovery in the last century. And this is a continued area of investigation in so many other tumor types because all the cells have HER receptors on the cell surface. And to the extent that we can give the right drug to the right patient, then we’re going to make a difference. And that’s why we’re now seeing approvals in gastric cancer, now even in colon cancer with the trastuzumab/tucatinib story in HER2-amplified colon cancer. And we hope to see many more other types of tumors that can be targeted with HER2-targeted therapy.

DR LOVE: So we’re going to be talking a lot about T-DXd, of course, but also, as long we’re talking about sort of the HER and EGFR axis, I wanted to ask you about HER3. We did a program with Dr Pasi Janne recently, where I think I picked up a little more about this, but still trying to figure it out. So you have an antibody drug conjugate, patritumab, that targets HER3. In your mind, how does that work?

DR LI: Yeah. So HER3 is another neighbor in the family. So you’ve got EGFR, HER1, and you’ve got HER2, and HER3 and HER4 are next door neighbors who actually cross-talk with HER1 and HER2. So we call that process dimerization, when the 2 receptors bind together and they send signals to one another, reinforce the signaling down to the nucleus.

Often to send that signal, the way it’s worked out is that the HER receptors need to talk to each other to send that signal. You can’t send that signal alone. So it’s either HER2 would talk to HER2. Multiple HER2 receptors get together and send a joint signal to the nucleus to send — to tell them to grow, and that’s called homodimerization. And then if you have EGFR or HER1, it’s the same thing. And then HER1 will get together with HER2 or HER3, and then they decide — the 2 receptors decide to send a joint signal to the nucleus to grow; that’s called heterodimerization. And it is really the HER1 and HER3 heterodimerization that is actually required for EGFR mutation signaling. That lends itself a target for HER3 antibody drug conjugate.

And remember, I said that when you activate, through dimerization that is, you actually ubiquitinate and internalize. So those receptors get sucked into the cell as you do so.

So when you bind the cancer cell with a HER3 antibody drug conjugate, like patritumab deruxtecan, or HER3-DXd as they call it, then it binds to the HER1 and HER3 dimer. And that dimer will suck the drug into the cell. And that dimer is going to be more active if you’re EGFR-mutated to begin with. So that the HER3 will latch onto the HER3 and EGFR dimer and then it gets sucked into the cell and releases the smart bomb and kills the cell in a totally different mechanism to EGFR inhibitors which is the lock and key approach.

And so, rather than switching it off with the lock and key, you’re really just throwing a bomb into the cell and then blows it up with the chemotherapy. Very different mechanism of action and, therefore, it has the potential to overcome all types of resistance mechanisms to TKIs.

DR LOVE: So within that model, what’s your vision about how pertuzumab works in breast cancer? For that matter, how trastuzumab works?

DR LI: So trastuzumab is a naked antibody without the warhead. So it doesn’t have the smart bomb. So it’s a missile, but it doesn’t have the bomb. So it’s a very different molecule.

What it does is, it’s got multiple mechanism of action. And to this day, I don’t think anyone’s really crystal clear in terms of how trastuzumab has produced its wonders in so many patients. It certainly signal inhibition capability, so it basically binds to the HER2 protein and tries to tell it to stop. So it’s antibody inhibition. So it sends a rather weak signal on its own to reduce the signal for growth. But more importantly, it actually elicits an immune response, ADCC, and that’s certainly part of the innate immune system but it’s antibody-dependent in a cell-mediated cellular cytotoxicity. So that will localize the enemy with HER2, and then elicit these antibody responses that are cell-mediated. And then we can really elicit a long-term response through that ADCC mechanism in the long term.

And that’s thought to be the main mechanism of action for these 10-year survivors of metastatic breast cancer on trastuzumab, 10-15 years, and then you get into the Neil Love question of when do you stop? When do you continue? And you’re really relying on an immune response in that sense. It’s not so much the chemo that’s doing the work.

And certainly there’s synergy with chemo as well, which may actually elicit that immune response. And that’s why in breast cancer you have to give it with chemotherapy. A naked antibody alone, it doesn’t cut it in terms of response rate and survival.

So in lung cancer, the early days of trastuzumab-based clinical trials have been done since the trastuzumab revolution. But all 6 trials were negative, including a randomized Phase II trial that was published by Dr Gatzemeier in Annals of Oncology. This was early days, beginning of the century. And it was all negative. But the patient selection was very different. It was really looking at HER2 expression. But we know that expression tends to be low in lung cancer, not has high in breast. It’s not as frequent to get amplification. And in the subgroup analysis, they saw a signal in HER2-amplified cancer, but they were only like 6 patients so it’s had to draw any conclusions.

So it’s the difference in molecular epidemiology that actually pretty much guaranteed a failure of these trials even though they were done in good faith, replicating the breast story into lung but the epidemiology was different and, therefore the trial design was flawed and they failed. But I do see a role in trastuzumab and pertuzumab in the longer-term future when it’s done in combination with other medicines.

So the trastuzumab/pertuzumab story is certainly the CLEOPATRA trial a decade ago, and then many other improvements since, have been the mainstay of therapy even this day in first-line metastatic breast cancer and a HER2-targeted therapy. But in lung cancer, we’ve seen the My pathway study showing some response, in the 10- to 20% range, with trastuzumab/pertuzumab. Not enough to do the serious damage on HER2-mutants. But we are seeing some signal of activity.

So I believe it’s the weak effect of those antibodies on inhibition and then eliciting an immune response that may still be targeting. And I think in the long-term you really have to learn from the breast cancer docs — you give it with chemo to elicit the immune response and that may actually give more, better long-term outcomes. And there are trials ongoing. Certainly one was already published last year with Dr Julien Mazières in the JCO on giving trastuzumab/pertuzumab and a taxane chemotherapy. And it showed just under 30% response rate, which is better. It’s an improvement compared to naked antibody approaches. And in the future I believe there may be more combinations that could be exploited to render a more durable response.

I’m not sure about the tucatinib combination in lung cancer. We’ve certainly tried that with afatinib and lapatinib and it hasn’t really panned out very well. But the jury’s still out. And certainly you can leverage the same rationale about the immune response in antibody drug conjugates, which all contain trastuzumab. So there’s certainly an area of investigation combining them with immune checkpoint. Perhaps the innate immune system can have an interplay with the adaptive immunity of T-cells with the IO checkpoint inhibitors combined.

And there’s a body of preclinical work showing that ADCs can actually prime the adaptive immunity and render itself to better response. This is an area of my investigation as well, especially after giving ADCs, can you mount an immune response? Turning a cold tumor to hot, so to speak, and then render them respond? Knowing that these tumors don’t classically respond, but maybe you can turn that into a hot tumor and then respond.

I’ve actually done that in some of my patients and I’m very pleased that some of my patients are having long-term response to immune checkpoints with HER2 mutation, but done in the sequence of doing a bit of ADC priming and then giving immune checkpoints later on. And some of them are just responding for 5+ years and disease-free. They were on the verge of death initially.

But I’m not replicating it in every patient. So certainly, more translational research needs to be done and a lot of thought process needs to put in, in terms of clinical trial design to maximize these immune effects of trastuzumab-related targeted therapy.

DR LOVE: You hear about ADCs and IOs in different situations, like in bladder cancer, they’re all excited about ADC, enfortumab vedotin plus IO. They think that’s going to be… and it kind of makes sense because in a way chemo plus IO, which lung cancer really defined that.

Current and emerging roles of antibody-drug conjugates in lung cancer

DR LOVE: Let’s talk a little bit about what we know about ADCs in lung cancer. You mention T-DM1, and I remember there was a paper, I think Greg Riely was involved with it, and I didn’t think it was all that impressive. But did they see objective responses with T-DM1?

DR LI: So there were actually quite a few papers on this. Firstly, the international trial of T-DM1 wasn’t that impressive. It was published in Clinical Cancer Research with Solange Peters and Tom Stinchcombe and many others. And it showed a 0% response in HER2 2+ and a very low response rate in HER2 3, but maybe a signal in these amplified cases with T-DM1.

So that was the paper. It was a negative study but maybe some signal-finding that warrant further investigation.

The Japanese trial, published by Dr Hoda, et al, in Journal of Thoracic Oncology, also around 2018 also was pretty negative, but it was kind of — the selection was all-comers. It was expression, amplification, mutation. The mutants responded, but the response rate in total was pretty low because it was a mixed bag, and it was small numbers to begin with. So hard to draw conclusions.

And then our investigator-sponsored trial at Memorial Sloan Kettering, it’s a cohort in the basket trial of T-DM1, which I had led. And, in fact, that was a project that served the basis of my fellowship training, going from Australia to New York at Memorial Sloan Kettering, and that was translating the breast cancer discovery into lung cancer. And that was also the basis for my fellowship, basis for the ASCO Young Investigator Award at the time, and that cohort in HER2-mutants actually was read positive. So we had a 44% response rate. We had a median progression-free survival of 5 months. And this was in heavily pretreated patient populations, often third-, fourth-, fifth-line of therapy, with prior chemo, prior HER2-targeted therapy and prior IO. So these patients really had nothing left to give, except maybe some chemotherapy like vinorelbine or something.

So that’s thought to be an encouraging signal. But small numbers. And that’s what led the subsequent investigations and investment into this HER2-mutant patient population, a rather obscure field of thoracic oncology at the time, but now it’s certainly created a new paradigm.

DR LOVE: So let’s talk a little bit about T-DXd. First, can you talk about the way you maybe explain to your fellows sort of what it is? How it’s built? The double payload delivery. Why it’s different than other ADCs we’ve seen?

DR LI: So this is a new antibody drug conjugate compared to T-DM1. It’s the same trastuzumab backbone but it has a different warhead. And it’s DXd, in terms of DM-1. And DXd is irinotecan-likeness, it’s a substance, but it’s much more potent. It’s a topoisomerase-1 inhibitor. And it’s certainly much more potent than DM-1 and also, we can pack a lot more warheads into the same missile. So the so-called drug-to-antibody ratio is twice, compared to T-DM1. You can just pack a lot of bombs in the smart bomb per molecule.

And then the linker linking the warhead to the missile is also a slightly different linker in terms of engineering. So it is a cleavable linker rather than non-cleavable. Cleavable meaning that it can easily break off and then release the bomb with ease within the cell. And that also has membrane permeability so the drugs — the toxins can penetrate from 1 cell to the neighboring cell to the neighboring cell rather than being confined within the cell membrane. So it can penetrate the membrane and kill neighboring cells. So we called that so-called bystander effect. And that may be responsible for some of the pneumonitis and interstitial lung disease that we can see with trastuzumab deruxtecan. But it certainly has given us more advantage in terms of killing HER2-mutant tumors. Even with some tumor heterogeneity it can still go around and kill neighboring cancer cells.

So with all those differences in bioengineering, it is certainly a more potent ADC against this target. And we have seen activity in the T-DM1-resistant, HER2-positive breast cancers initially. And now, certainly the DESTINY-Breast03 and other trials have shown that it’s superior to T-DM1 even in the second-line setting.

So similarly in lung cancer, we did the initial Phase I trial in HER2-mutant, knowing that borrowing from the T-DM1 story, which I had the opportunity of leading together, in collaboration with Greg Riely and many others at Memorial Sloan Kettering, just borrowing that concept and then to develop T-DXd or in HER2 or trastuzumab deruxtecan. We were able to hone in on the HER2-mutants. And in the Phase I trial that we published in Cancer Discovery, this is Tsurutani, et al, in 2020, Cancer Discovery, and we showed that all-comers, it was only a 28% response rate. But in the HER2-mutant lung cancer cohort, we had a more than 70% response rate. Small numbers but very encouraging signal when you hone in on HER2-mutants with this more potent ADC.

And that led to the DESTINY-Lung01 trial which we did as Phase II, and this was launched in 2018. But we’ve published that earlier last year, just about a year ago, in The New England Journal of Medicine. And as Dr LaRusso had pointed out in the ASCO Plenary discussion, that case in the waterfall, the one with the complete response on the right-hand side, actually was HER2-0 in terms of protein expression.

So it actually has nothing to do with the protein. Back to the photocopier, making lots of copies of HER2. It’s got nothing to do with that. It’s the function of HER2 mutation that sucks the drug into the cell that made a difference. So they had to be activated and mutated. And that trial, then of course a little bit of boast — tweaking the 6.4 mg/kg was thought to be too high. We did DESTINY-Lung02, which compared that with the lower dose and found that the lower dose was just as effective, but less toxic. And that was subsequently served as the basis for FDA approval just a few months ago, in August 2022.

And so that’s the story, going back about a decade or so, behind the use of antibody drug conjugates targeting HER2-mutant lung cancer. It’s the first ADC to be approved in lung cancer, but we are hoping that this creates a  new paradigm of a new class of drug that could be helpful for patients with lung cancer.

DR LOVE: I thought I understood ADCs until T-DXd came along. So first of all, you mention DESTINY-03 Breast, where way better outcome than T-DM1. So 1 ADC versus another, huge difference. And the HER2-low, Dr Modi presented at ASCO, again you think a targeted therapy, now all of a sudden, well wait a minute, I’m not sure what the target is. I think I’ve seen some data even with responses in HER2-0. You have the TROP ADCs that, even though not everybody has TROP, everybody seems to respond. So, I don’t know. I think there’s a lot of confusion, at least in my mind a little bit, about sort of what these agents are and sort of where they’re heading.

DR LI: Yeah, I think it’s confusing for investigators and scientists as well. So we’re all learning in unchartered waters. There are 2 drug development pathways for ADCs. One is the precision pathway, which is personally my preferred way of developing drugs. You really want to bring the best drug to the right patient for maximal outcomes. But that’s usually a small slice of the pie, a small percentage of patients, and that approach has its pros and cons.

The HER2-mutant story is the prototype for that. And I hope to develop more for HER2 amplification and many other targets down the track.

But there’s another drug development pathway for ADCs and that’s using it as just a chemotherapy, so it’s given to everybody like nab paclitaxel. So you give it to everybody. And you’re not going to aim for a precision medicine type of response rate or duration of response, but you’re going to aim for like a docetaxel-like response, maybe slightly better, and then they can get a drug approval in the later-line setting.

And that’s a strategy that’s currently being developed for TROP2 and there’s CEACAM and many other ADCs that are emerging in the lung cancer and other solid tumor space. And they don’t necessarily need a companion diagnostic or a biomarker to select patients because they’re just going to give you like a 20% response rate. And then in the refractory setting, arguably it’s better than nothing. And even though the duration of response may not be super high but, overall, it’s something. And there may still be a drug development and approval strategy in that population.

Personally, it’s rather unsatisfactory to do it like that because it feels like you’re going backwards. You’ve got all these engineering, and then you’re going back to the old days of chemo, a more expansive version of chemo, though certainly, I can understand the company pushed to do so. But as an oncologist, I’d like to really see a lot more precision, especially in this day and age of technological transformation, to do some work, honing in on the biomarkers of response. Really find out, it’s got to be more than just IHC and mutation. It is a lot more ADC biomarkers that need to be uncovered to really select the right patients who may respond to the right ADC.

So I think the jury is still out in terms of where the future of ADCs will be. But certainly, I’m hoping for a more precision drug development pathway.

DR LOVE: Yeah, and I think that idea also ties into — I was kind of confused when I started to see the data looking at dose, that you just referred to in lung cancer. The study where you increase the dose; you just saw more toxicity and not greater efficacy.

We had a couple of cases presented at San Antonio breast cancer meeting by a doc, actually from New York, in the community, where he gave T-DXd to HER2-low patients who were like 90 and 91 years old. And he decided to start at a low dose and escalate them up, although I don’t know even know if that’s been done. But again, it sounds like you said, a little bit like chemo and how it’s been approached. Any thoughts about that?

DR LI: Yeah, it’s a good question. In fact, we’ve looked at the data closely in the Phase I dose escalation study. So, in that study, we started with low dose and we then dose escalated, 4.4, 4.5, 6.4 and even 7.4 milligram per kilogram. And we saw that the highest dose was too much, there was a lot of toxicity. We had patients, you’ve got 1 dose — the big, strong, muscular men who get 1 dose and they’re just lying in bed for the whole week and they’re having constant diarrhea and so forth. It’s just not compatible.

So we pushed it back and they dose de-escalated. And then in the Phase I trial, we came back with the 2 recommended Phase II trials which led to the confusion — there was a 5.4 and 6.4. And there were arguments on both sides. So both were RP2D. In breast cancer they’re a lot more exquisite in terms of chemosensitivity. So a 5.4 dose was pursued because they thought some investigators and the drug developers had thought that you don’t need too much dosing for breast so you can get away with 5.4. And that was right.

I had always argued for 5.4, but there’s also another in lung. But there’s a school of thought that lung cancers are more chemo-refractory. They’re not chemo-sensitive and, therefore, you need a stronger dose. And there’s a no pain, no gain kind of concept.

So 6.4 was pursued, but evidently, the DESTINY-Lung02 had proven that 5.4 is just as good and less toxic.

So I think you do get into like a plateau effects. There is a dose response relationship, just like chemotherapy. But because it’s through a targeted delivery mechanism, it’s not always the more the merrier. So you’re getting to a 5.4, even a 4.4 — a lot of my patients in the early days of drug development have been dose reduced to 4.4 and they’re still responding today, for many years. So I think a lot of may be eliciting an immune response long-term and the dose really doesn’t matter when you get to that point.

So, yeah, it is treated like a chemo but it’s not exactly the same.

DR LOVE: Well, the other thing, and when I first started hearing this from the breast people, I’m not so sure in terms of lung and GI, but I started to hear about alopecia, GI effects. I think the PI actually says you should give preemptive GI meds for that. What have you observed in lung cancer in terms of those issues?

DR LI: It is certainly, it’s variable per patient. I’ve had patients wo have no problems and others who just, like they’re really miserable. And it mirrors some chemotherapies and it can produce really horrible side effects and others sort of just get along — go about with life.

So I tend to use palonosetron and dexamethasone as a chemotherapy premedication, just like if I’m to give carboplatin. So they’re moderately immunogenic. And in some, I don’t routinely give fosaprepitant, but you can. and when you do that you’re going to constipate some patients. So that could offset some of the diarrhea effects. So I don’t tend to give loperamide like you do for neratinib and some TKIs, afatinib and other TKIs. I don’t tend to do that.

I find that most patients wouldn’t need it if they get enough antiemetics. But certainly diarrhea is possible and we have to account for that and manage it with supportive medicines at the earliest instance.

DR LOVE: So you talked about the first-line trial that you’re doing, and earlier you were talking about the IPASS study, which again for people new to oncology, that was really the first study that compared targeted therapy to chemo. Then they did stuff like that in ALK. Then they stopped doing that. And if people had response rates of 50-, 60%, people just started using it first-line like in RET, et cetera.

But again, this is a little bit different. So globally, when you look at T-DXd and HER2-mutant lung cancer, first of all, what are you seeing in terms of response rate and also duration of response? And how does that compare to what you see with osimertinib or the ALK inhibitors?

DR LI: The mechanism of action is quite different compared to the EGFR-TKIs. So there are some corollary, some areas of similarity, but overall, I don’t think they can be judged the same way. And it is kind of like a smart targeted chemotherapy, so I’d see it like a precision chemotherapy, so to speak.

And the response rates have been about 50- to 60%. And interestingly, I’m seeing some immune effects. Some of my patients, even treated off-protocol, some of them respond very quickly in the first scan and get a partial response. But some of my patients actually get stable disease for a year and a half and then they start to get the partial response. And I just had a patient, 2 years out, and he’s got a complete response.

DR LOVE: Wow.

DR LI: So those are not the same as classic targeted therapy mechanisms. You take the TKI and then you have this resurrection effect from the ICU that you tend to see with this lock and key approach of tyrosine kinase inhibitors. This is more chemotherapy. It’s a slow, gradual response. But then there’s this immune component that can come later and then you get this melt down of tumors.

So I think they’re different. So I don’t just look at the response rate. You have to look at the full picture. And the full picture is certainly looking promising. In a refractory setting, you’re seeing a median PFS or progression-free survival, of about 8 months. You get a median duration response of 9 months. But this is in heavily pretreated, refractory setting. If you bring it to first-line, just like what we saw with DESTINY-Breast03, you may get much better outcomes in terms of progression-free survival. And interestingly, in the Breast03 study, we also saw less toxicity if you give it upfront rather than give it really later on.

There were no treatment-related deaths. The toxicities were much better managed. So if you give it early line, of course you learn how to manage the toxicities over time. So that’s also made a difference.

But we are hoping in DESTINY-Lung01, we’re going to see some similar improvements in outcomes if you give it first-line rather than in the chemo-refractory setting. And if it’s a smart chemotherapy, you really want to give it early on rather than when the drug is getting resistant.

So there’s some hope that this would be superior to even chemotherapy and pembrolizumab in the first-line setting, if we choose the right patient population. But the jury’s still out. We still have to complete the study.

DR LOVE: And in that study, first of all, do you allow pembro alone for high PD-1 or they have to get pembro/chemo?

DR LI: It’s pembro/chemo, just to keep it homogeneous so you don’t have too many biases. And also, there’s a school of thought in HER2-mutant, they’re all HER2-mutant. So regardless of PD-L1 expression, maybe the response is going to be slightly lower with single-agent IO, and, therefore, we give chemo/IO. That’s the thought process at trial design. There’s a lot of debate about this. And I could easily see this going into a Neil Love Point-Counterpoint session when you have investigators debating. There’s no hard answer to this yet.

DR LOVE: And in the T-DXd arm, it’s just T-DXd? Not with an IO? There’s not a third arm? So just straightforward? Yeah, I’m sure you debated that, too.

DR LI: Yeah. We’d also discussed about the IO combo, but there’s a school of thought to keep the trial simple; not to ask too many questions in 1 trial. Just keep in simple. Get a straight answer. Otherwise, you get too many confounders. And it’s also a difficult trial to accrue because it’s 2%. Not everyone gets HER2 testing routinely. And therefore, if you get too many arms it’s going to be twice as long to complete.

Considerations for clinical trial participation and improving access to care for patients with cancer

DR LOVE: Absolutely. That’ll be really interesting. When do you think we’re going to get a read-out on that?

DR LI: So I presented the trial design last year in ASCO. That’s just without — just that reporting that we started the trial, we started enrolling and it’s going well, and the design. But I’m hoping maybe, preliminary, maybe next year. I don’t know for sure. But it really depends. It’s also difficult to accrue in the first-line setting because patients, when they’re diagnosed, they’re understandably anxious. They want to get onto a treatment quickly. And the trial, you always have to test for the mutation. And not everyone gets upfront HER2 testing and it takes a long time to get NGS back.

So at Memorial Sloan Kettering, we’ve rolled out a Reflex PCR testing for all patient with non-small cell lung cancer, just to catch those 2% of patients who may benefit. And that takes a few days. But if you do NGS,or liquid biopsies, or at MSK, we’ve used Memorial Sloan Kettering-IMPACT, that all takes a couple of weeks. And then you’ve got to identify that. Then you start the consent process. Then you have trial eligibility screening. And then enrollment. And then treatment. So that is all time. So it’s sometimes hard for patients to wait that long and to get onto the trial. So it’s somewhat difficult from that. There are some logistical hurdles.

But as we were talking earlier with the Bloomberg New Economy International Cancer Coalition, we’re really trying to work together across stakeholders to simply clinical trials. Make it a less cumbersome process for patients to get on. To provide access more to community — patients in the community and community oncologists through maybe a decentralized trial model or a hub-and-spoke model that we published in Nature Medicine, we described it. Through that collaborative model we may be able to bring innovation to patients, not only in the academic centers, but across the board in community. And that serves that disparity concept that you have talked about earlier on.

Disparity and equity, from my standpoint, is not only for social justice, but it’s also — which is important on its own. But in my world of curing cancer, it’s absolutely critical to engage all patients, then you can accelerate the accrual. My T-DXd first-line trial, if done in a really diversified, collaborative manner and bringing it to the patient, we can accrue like wildfire. You asked me, when’s the read-out? With that approach, maybe this ASCO, if we can do it properly. With true international collaboration we’re going to accelerate the timeline of drug discovery and approvals. And you fast-track that to all trials, you’re going to see the cure for cancer. So it’ll keep you very busy with the podcasts. But it’s really something that I can see happening if we get it right through collaboration.

Because my patients, many of my patients with metastatic lung cancer, thought to be a death sentence just not that many years ago, they’re living 5, 10 years out. I’m seeing these patients who have complete response. Many are in NED. You do a scan, there’s no evidence of disease. They had brain and spine metastasis to begin with. But I just monitor some patients with a scan every 3 to 6 months. And they’re not cured because they can come back, the metastatic disease, from the outset. But they get on with their lives.

So that’s not a complete cure, but I think we’re getting closer and closer to the cure for cancer. But we have to do the clinical trials right. And that’s the issue that we’re currently encountering. What if everyone in the world with HER2 mutations will get a notification on their app, and it’s optional — we’ve got to protect their privacy and confidentiality and everything — but there are ways to do that. You get an app and you tell them where are the HER2 trials, first-line, second-line. and then we work with the local oncologist to bring that medicine to the patient.

So we published a study a few years in JAMA Oncology at Memorial Sloan Kettering. We already showed that more than, about three-quarters of patients — in some studies quoting higher figure — are treated in the community. Patients want to get their care close to home. They don’t want to travel thousands of miles and jump through hoops and stay in a hotel. And wait in the waiting room of academic centers. They want to continue their life. So they want to be treated locally. But the trials are often not available because it’s so complicated. And community docs are busy. They’re busy enough treating patients. So then you get suboptimal care because we know that clinical trials are widely recognized as the best care of patients with cancer, but we’re not making them easily accessible and available to patients.

So by doing that, leveraging the community docs, which your enterprise has vast networks of access to community docs, if we can rev them up and then we change the equation and change the rules of clinical trials, make them to be the key players, the leaders on clinical trials, with support. And there’s a lot of industry funding in oncology research and we can easily hire an NP or a PA to support clinical trials in community practices, to support the community docs. Then together, through a collaborative model, we can really accelerate these trials to completion and, therefore, the cure of cancer.

DR LOVE: So I was telling you about this disparity presentation we did. One of the things that we did, because every now and then we also do programs where we interview patients, and I showed a short clip from an interview I did of a patient with myeloma who’d been on the bispecific antibody teclistamab for a year and a half, doing great, responding. No symptoms. And the day that I interviewed him was the day it got approved. And he’d been on it for a year and a half. Not only that, it was approved as fifth-line therapy; he got it as like second-line therapy. So it was just a classic example of the advantage of being in clinical trials.

Anyhow, getting back to T-DXd, I think you referred to the fact that — one thing I was curious about, and one of the reasons I asked you — I know it’s different than ALK and EGFR, but also, from the point of view of docs in practice it’s different. To me, it looks like the responses aren’t as long. You said 9 months, 8 or 9 months, which, it sounds shorter than like what you see with EGFR and ALK. Is that the case?

DR LI: Yes. It depends on the chemo-sensitivity of the disease. So in breast cancer, you’re seeing a bit longer than that. and, also, it’s longer if you give it upfront in the first- or second-line setting rather than fourth- and fifth-line. So I think those numbers can change, depending on the clinical context and we’re hoping that DESTINY-Lung04, which is the first-line trial, will produce superior outcomes. But the jury’s still out. And it’s really a chemotherapy delivery system. So it’s a bit different to the tyrosine kinase.

So I can’t really equate one with the other. You can probably borrow in an EGFR/ALK/ROS1/RET, these are very similar in terms of their action — mechanism of action. But ADCS are very different. And, therefore, randomized, controlled trials still play an important role.

Management of trastuzumab deruxtecan-associated interstitial lung disease

DR LOVE: All right. So let’s tackle the elephant in the room, ILD. And of course, we talk about this all the time with the breast and the GI people. And our discussion in the lung people are a little different. I don’t know whether it’s because there’s so much pulmonary issues already with the typical lung cancer patient because they smoked and COPD, etc. But I don’t know, I kind of get the feeling that the lung people aren’t as concerned about it. But anyhow, I know you’ve been very involved. We did a whole program where we worked with a pulmonologist, trying to dissect out the whole issue of ILD.

So can you talk about what’s been seen in the lung cancer trials in terms of ILD in comparison to what’s been seen with breast and GI? And for practical purposes, right now — and incidentally, you’ve got your first-line trial. But we have our way of doing stuff is we just go out to 25 investigators and say, well, let’s put the reimbursement aside, what would you like to do? And I’ll tell you that half of your colleagues right now would like to give T-DXd first-line.

DR LI: Yeah. Yeah. It shows the historical perspective of thoracic oncology. We favor targeted therapy over chemo. So that’s almost a dogmatic approach to things. You’ve got a targeted therapy. You’ve got a target. Then that’s the way to go.

And personally, I feel the same way but I just need to put it to a trial to prove it so that we do it right because you can get it wrong. And it’s a different mechanism of action to the pill, tyrosine kinase inhibitors.

In terms of ILD, it’s not benign. It could certainly be fatal. Our DESTINY-Lung01 trials had reported the fatality. So did most T-DXd trials, except DESTINY-Breast03, which gave it early line, with an extensive education and early intervention program that actually made the difference. So, we are hoping to improve upon that statistic.

In the DESTINY-Lung01 trial, we’re seeing about a 13% rate of pneumonitis and ILD. So, largely, you can say 10- to 15% is sort of the ballpark. But we’re hoping to reduce that number. Most of them are benign — are pretty mild, Grade 1 and Grade 2, about half of them. But some are serious. So I would say that 75%, we can say are manageable. They’re usually treatable, lower grade. But there is this 25% of ILD that is potentially serious and you might need hospitalization and you might need intravenous steroids. And if you intervene early, it’s good. But if late, it could be fatal.

Thoracic oncologists are kind of used to doing this with IO-related/immune-related pneumonitis. It was once upon a time very scary as well. And we had lots of education workshops in CME how to manage immune checkpoint. Lots of guidelines being published. And I had participated in some of them in the early days. And I can still remember, back then there was a lot fear. But now it’s bread and butter for us as thoracic oncologists. And we’re used to people who are coming with oxygen tanks, with COPD and all that. And you know that these patients are higher risk. But we still give immune checkpoint inhibitors. So we’re kind of accustomed to that sort of high-risk population and managing it.

In a way, it’s not that much different in terms of T-DXd ILD management. It’s early recognition and early intervention with steroids and oxygen. So, if we get that — we learned how to do that with the immune checkpoint pneumonitis, then the thoracic oncology docs are a little bit more comfortable dealing with it in the ADC context.

DR LOVE: So I think I figured out the algorithm to detect ILD in the breast cancer people. I think. First of all, their approach is, if you pick it up when it’s asymptomatic, then theoretically you can rechallenge. If you pick them up when they have symptoms, they’re done. Does that algorithm apply to lung cancer?

DR LI: Yeah, generally that’s the same algorithm that we’re using across the DESTINY trials. And we recently published an algorithm in Cancer Treatment Reviews with Sandra Swain as first author. So that sort of serves as some guide, just like the early days of IR-pneumonitis. Which is actually quite similar. In IR-pneumonitis, you rechallenge when they’re asymptomatic if you patches or you have very mild symptoms. But if you’re hospitalized, if it’s got severe complications, they you really think twice about rechallenging.

DR LOVE: So it’s not necessarily that easy to read those kind of papers and figure out like what you’re supposed to do. So, the other issue is imaging. So again, the breast people right now are saying, we’re trying to get imaging every 8 or 9 weeks. Maybe for a long time. Even if they don’t need the imaging, because of their metastatic disease, because we want to pick this up asymptomatic. Again, is that kind of the way you see it going in lung?

DR LI: On the trial, we’re seeing it every 6 weeks with imaging. That’s mandated by the trial for that very reason, of picking it up. But in standard practice, I tend to do it every — initially, perhaps every 2 to 3 cycles. But over time, every 3 to 4 is perfectly appropriate. And if the patient is asymptomatic, they’ve got their pulse oximeter at home and I’m checking everything, I don’t think that if too frequent scanning is a bit of a burden for patients. And I don’t think there’s great evidence saying that if you just hold the drug at the earliest onset of a little patch that are non-specific, it could be that you just had a flu or you had COVID, and you don’t know; you can’t work it out on the CT scan. Then what do you do?

So there’s a lot of unknowns in that context. I think on clinical trials you’ve got to be very vigilant. But in standard practice, I think it has to be a  bit more personalized. I wouldn’t be as dogmatic. Yes, you do need the CT monitoring every few cycles, whether it’s every 2 or every 4, I think it depends on the patient.

DR LOVE: So I know if we present a patient at one of our breast cancer programs who already has some kind of lung problem, they get very, very nervous about using T-DXd because they don’t see it that much. Most breast cancer patients don’t have preexisting lung problems. Yet, most of your patients do. So again, how does the trial handle prior COPD and other cardiopulmonary problems? And clinically, what do you when you want to use — you have a HER2-mutant situation, but yet the patient’s got maybe even symptomatic COPD.

DR LI: So those cases, it’s a risk-benefit ratio discussion with the patient. The same could apply with IO checkpoint, which can cause pneumonitis in a fair number of patients. And if they have COPD and they’re on oxygen, and they have like one lung, do you give it or do you just let them go? I mean that is a question that we have to individually discuss with the patient.

But in my experience, it’s not an absolute contraindication. And there are instances where the cancer is going to imminently kill you in the next few weeks unless you do something about it. And you’ve got this drug, you’ve got a shot. It’s a double-edged sword. Do you give it a shot or do you just let yourself go? I mean the patient needs to make that decision, together with the oncologist. But I don’t see this is as an absolute contraindication, particularly coming from the lung cancer perspective where they all have some lung conditions.

So I think that speaks to the uncertainty and the experience from the big, different subspecialities. And it’s again a reflection of different silos and different ways of thinking.

In breast, their lungs are healthy. Breast cancer community, and the earliest incidence of any lung condition, they get frightened about it. And it’s the uncertainties can produce a lot of fear, not only in patients, but among oncologists as well.

The thoracic oncologists are much used to doing this. So there’s a lot more comfort level. And I think you can learn from one specialty to another. And that to give a plug to community oncologist who see different — a variety of different clinical context and tumor types, they can get a better appreciation of what’s clinically important. And some of them are clinically significant, but many of those concerns are not.

So I think we need to take a step back and hear what, like what you’re doing, hearing from different points of view, and then take a balanced view of this.

But my approach to T-DXd pneumonitis is very similar to my approach to IR checkpoints.

DR LOVE: You bring up the general medical oncologists who’s hearing from the Gi people, they’re hearing from the lung people, hearing from the breast. One of the things they hear from the GI people that they don’t hear from you and they don’t hear from the breast, is the — you can ask Dr Janjigian, they’re all into repeating HER2. So if they’re HER2-positive, and then they’ll do another assay when the progress or something, and if they’re HER2-0 then they don’t use a HER2. Whereas with breast, once you’re HER2, they’ll just keep using it.

Again, any thoughts about retesting in lung cancer?

DR LI: So if it’s HER2-mutant, it’s like EGFR mutation, they tend to stay. So they don’t go away. It’s an oncogenic drive. It’s constitutively active. And I wouldn’t need to retest it for the sake of checking if it’s still there. But I would do the repeat biopsy and NGS just to look for resistance mechanism that may emerge and that may be targeted with new therapeutics.

So I do a lot of rebiopsy, more for our clinical research and clinical trial matching point of view, but not so much for the HER2 mutation testing. And with amplification, things can be a lot more heterogeneous. And it sounds like the loss of HER2 is a big thing in gastric, but not so much in breast.

DR LOVE: Yeah, that’s a great point though about the mutations. That’s really interesting.

Future of HER2-targeted therapy in lung cancer and beyond

DR LOVE: Let me ask you, do you envision in the future that maybe there’s going to be a PanTumor approval of say, T-DXd for just HER2-mutant disease in general regardless of the primary?

DR LI: Yeah, that’s a great question. And in fact, this is the concept for the PanTumor-01, the DESTINYPanTumor-01, which I had presented the design a couple of years ago at ASCO. And we just completed accrual. And the jury’s still out. We have to do some data analysis properly. And this is specifically looking at the HER2-mutant solid — different types of solid tumors.

This is a pretty simple design. It’s the same drug. You give it in an oncogene-selected patient population across multiple tumor types that are not lung cancer, because we’ve already gotten approval for lung. So we’re looking at biliary tract cancers and colorectal cancers and breast cancers that are HER2-mutant but not amplified or expressed, and ovarian cancers. And then we’re looking at responses in those. And this is a trial with a PanTumor approval in mind. And whether that’s going to pan out or not, we’ll have to give it a little bit of time. But I’m hoping to read out the results much sooner because we had completed accrual. So this is a very important study that could open up a new indication for patients with different, not just lung cancer, but many different types.

DR LOVE: Yeah, the oncologists are all over that. we actually had a generally medical oncologist who presented — we were doing a Gyn program — 2 cases of cancer of the endometrium. Now I didn’t ask if it was HER2-mutant or HER2-overexpressed, but the doc gave both patients T-DXd. One had a CR; the other one had a good PR. So I’ve got a feeling people are going to start doing that before your trial is reported. But it’d be great to get the data.

DR LI: Yeah. Yeah. That’s very encouraging to hear. And certainly I’m going to feed that back to my co-investigators on this study to really try to push this paradigm forward. It’s a little bit difficult to tease out because now you have HER2-low breast cancer approval. So, is it driven by the mutation or is it driven by the low protein? And sometimes a lot of biomarker analysis needs to be done. But this is an important, new concept and it could deliver a breakthrough drug for many, many patients in need. And you just cited some examples of clearly, some lives being saved.

The DESTINY-Lung04 is still a little bit of time away. And this is an international trial that we’re doing across North America, in Latin America, Asia and Europe. So that’s what it takes to accrue to a study of this size. And this is a rare tumor population, arguably not commonly tested. Most oncology practices do not test for HER2 unless it’s part of a broad NGS panel, which takes time to get results.

DR LOVE: One of the things I like about this is that you could really get equipoise about either arm. I don’t see a problem randomizing somebody on a trial like this. I think it’s really hard to predict which one’s going to be better, for me. I think most people would be fine with it.

DR LI: Yeah, exactly. Because if you get chemo, we could still give them T-DXd later on. So you can get the best of both worlds. And most patients are actually very happy to do this. And if this could be — I mean the trial is already started, but I’m just using it as an example of such trials, and many of such trials are going to happen I the future. If we can broaden access to community oncologists and make trials really a reality, not just in the academic centers where you’re only seeing a quarter of patients, but broaden it to the three quarters of patients who are seen in the community, that’s going to lead to a sea change in terms of the pace of breakthroughs. And also the ability to collaborate internationally. Because if I can collaborate with someone in Nebraska in community oncology, and we can through telemedicine and through Zoom, you can connect to all corners of the world if there’s a wi-fi. And we could definitely do a lot of the monitoring also remotely. You still have to have a very good local oncologist and that’s essential. No technology is going to be able to replace that. But you empower that oncologist with technology, with resources, we can really do a lot.

So I’m hoping that the Neil Love network of oncologists could also be empowered to do their part and lead trials in that regard.