Anti-HER3 Antibodies May Inhibit Growth of Tumors Harboring NRG1 Fusions - Episode 1

Use of the Anti-HER3 Antibody Seribantumab in Preclinical Models

,

Experts review a preclinical study of the anti-HER3 monoclonal antibody seribantumab, used in cell lines and mouse models of solid tumors.

Transcript:

D. Ross Camidge, MD, PhD: Hello. Welcome to CancerNetwork® Between the Lines. Today’s featured article in this journal is “Anti-HER3 Monoclonal Antibody Seribantumab Effectively Inhibits Growth of Patient-Derived and Isogenic Cell Line and Xenograft Models With Oncogenic NRG1 Fusions.” I’m Ross Camidge from the University of Colorado. Joining me today is my friend and colleague Dr Alex Spira from Virginia Cancer Specialists. We’ve got an academic and a community oncologist. Alex, it’s great to see you.

Alexander Spira, MD, PhD, FACP: It’s great to see you, Ross. Thanks for having me today.

D. Ross Camidge, MD, PhD: We’re going to jump in with this paper. I love the fact that when we were talking about this before, you said this is the information that they show at site-initiation visits, but we’re going to pull it apart a little. You may not have heard of NRG1 fusion. NRG1, neuregulin 1, is 1 of those ligands that binds to some of the HER [human epidermal growth factor receptor] family. In particular, it’s the ligand for HER3. The picture on the right-hand side of this slide, which is from the seribantumab website, isn’t a great picture because what happens with the fusion is this ligand is fused to something else that essentially tethers it to the cell membrane. Instead of it being released and floating around, suddenly it’s stuck there. What it can do is stimulate HER3 on the cells as an autocrine effect or on nearby cells in a paracrine effect. When HER3 is active, it partners with a number of different members of the family—mostly HER2, but it can also partner with EGFR and HER4—and causes signaling. That seems to drive some cancers. Alex, have you already heard of NRG1 fusions?

Alexander Spira, MD, PhD, FACP: I’ve heard of NRG1 fusions, and this is interesting to me for a couple of reasons. Seribantumab has been around for a while and has actually been repurposed. It was originally studied broadly, trying to look at HER3 overexpression. We actually did that study, previously known as MM-121, a long time ago and now it’s being rehashed in this setting. For me, the interesting thing is that as you think about this, we typically think of drugs and targets. But as we look at NRG drugs and NRG fusions, we’re targeting the HER3 receptor. We’re not directly targeting the NRG1, but we’re targeting its partner, HER3, which is a different way to think about typical targetable mutations. I’ve heard about it. One of the challenging things is that we’re looking at a very small percentage of cancer patients. Less than 1% of cancer patients have this targetable fusion, although it’s very interesting because like most targetable mutations, when you have it appears to be very active. If you can inhibit it, you get tumor regression and tumor suppression, at least in theory.

D. Ross Camidge, MD, PhD: There’s a little devil in the details there, but we can go into that.

Alexander Spira, MD, PhD, FACP: That’s certainly the case, but as you said, it’s incredibly rare as well. Why don’t you go into the devil of the details?

D. Ross Camidge, MD, PhD: Let’s do it. You’re right. It used to be called MM-121. It used to be a Merrimack Pharmaceuticals drug, and it was originally developed because HER3 was perceived to have a particular role in EGFR-mutant lung cancer. They were trying to interfere with it to see if that worked. It didn’t really pan out, although it was interesting that HER3 is potentially getting another life as a target for an antibody-drug conjugate. But let’s focus on seribantumab. This is how it’s supposed to work. You’ll give it to this particular cell line, and it’s engineered to have 1 of these NRG1 fusions. This is a western blot, and what you can see here is increasing doses of seribantumab directed against HER3 vs afatinib. This is our positive control. This is a pan-HER TKI [tyrosine kinase inhibitor], and in the red boxes, you can see phospho-HER3, active HER3, going down with increasing doses of each of the drugs. You can also see, because HER3 forms these heterodimers, that activation of phopho-HER2 is going down, activation of phospho-EGFR is going down, and activation of phospho-HER4 is going down. This is the drug doing what it says on the side of the can. That’s all this western blot is telling you.

Alexander Spira, MD, PhD, FACP: For me, there are 2 important things. When you think about this pathway, afatinib clearly has activity as well. It’s a direct pan-HER drug, but there’s activity. We can talk later about what the limitations are, but this is the first test. You do the first thing, and then you get down to signaling. It’s a very easy thing to do to see if you’re inhibiting the downstream phospho-proteins, which you are here. It’s an important first test of concept.

D. Ross Camidge, MD, PhD: This is a proof of mechanism that the drug does what it’s supposed to do. The next question is, does it matter? This is the information that they show at site-initiation visits. These are cell lines. These are engineered or derived from patients who have NRG1 fusions. You can see that there’s more than 1 NRG1 infusion—and we can go into that to some detail—that can be fused with different things. See how long that neck of the dinosaur is, which has got the ligand on the end there? All of this, even among this rare population, denotes some heterogeneity, and it’s going to be interesting to see how that plays out in the clinic. But if you ignore panel B, which is showing what the different fusions are, and you look in panel D, these are some cell lines.

This 1 is driven by NRG1 fusion, and you can see what we showed in terms of phospho-EGFR. In terms of growth, these inhibit the growth of these cells. It looks like afatinib seems to be more potent, so as you go up on the dose of afatinib, the growth goes down. The same thing happens with seribantumab, but of course this isn’t a cell line. Cell lines don’t have toxicity. If you look on the right-hand side, it’s 2 different NRG1 cell lines. It’s a different way of looking at the same thing. Instead of looking at the decreasing growth, what you’re doing is plotting the cell number. On the flatter line, there’s more inhibition of growth. Is inhibition of growth what we’re going to see in a patient? No, we want the cancer to shrink. This is proof of principle, but it doesn’t tell me the patient is going to feel better. How many drugs are truly cytostatic in the real world these days?

Alexander Spira, MD, PhD, FACP: This is a great proof of concept, as you say. The first thing I think of when I look at the slide is obviously that we’re talking about seribantumab, but what about afatinib? We can talk about some of the issues, but if you look at the neck of the dinosaur—I like that term, by the way—the first thing that I ask is, “Why haven’t we heard more about afatinib?” Afatinib looks like it’s doing a better job. You can get into some of the toxicities, and there are other issues there, but that’s the first thing I think about: the relative growth on these cell lines that we see in panel D. We can talk about that a little more, either now or later, as we go beyond just inhibiting growth. Are we talking about patients responding? That’s ultimately the important thing.

D. Ross Camidge, MD, PhD: If you look at this, it looks like afatinib is the key thing to watch. It shows the most effectiveness and then seribantumab is saying, “Hey, what about me? Can I be a contender?” You can see there’s a dose effect with seribantumab, and then you start to ask yourself, “If I choose a lower dose of seribantumab, it will look worse. If I choose a higher dose of seribantumab, it looks better. Why that dose of afatinib?” You start to pull this apart, and you ask, “Is that a dose that’s relevant to the exposures I would see in a patient?” It may be much higher. We showed inhibition of growth on the last slide, and I hinted that you really want to know if the cancer’s going to shrink. This is a step in the right direction. It’s still cell lines, but now we’re showing the breakdown of products of apoptosis. It goes up with dose, but it goes up with afatinib more than it does with seribantumab. What do you make of that?

Alexander Spira, MD, PhD, FACP: It’s a very similar concept, and it addresses a couple of things. As you think clinically about the doses and the concentrations they use, at least with afatinib, these are biologically relevant doses. To me, it looks like you’re getting more apoptosis with afatinib than you are at meaningful doses of seribantumab. There’s a question regarding the dosing of the monoclonal antibodies. It’s a little more challenging to do, but you clearly are causing apoptosis. Now we’re getting into more important end points: not only inhibiting growth but causing the cells to apoptose and die.

D. Ross Camidge, MD, PhD: This is a perfect graph to explore. Look at the concentrations. These are micromolar measurements, so 10 mm3: nobody is giving a drug at that dose to a person. Normally, we’re expecting drugs to be running around 10 nm3 to maybe a small number of 100 nm3. So, 10 mm3 is crushing the cell under the weight of the drug. Those results on the far right of the blue curve and far right of the red curve are not that encouraging, especially with seribantumab. That raises a few questions—you have to be at a level of dose that may not be clinically achievable.

I said you’ve got to shrink the tumor, and now in the preclinical setting, you go to a PDX [patient-derived xenograft] model. It’s probably implanted on the flank of a mouse. Just look at the curve down on the bottom left here. They let the tumor grow to a certain size, and then they start giving the drug. If you look at those curves, it’s about delaying the tumor growth. But what becomes interesting is if you look right in the bottom left-hand corner, you see the beginnings of shrinkage. This is the closest thing that we’re going to get to a patient with tumor shrinkage. If you look on the bottom right-hand side, zooming in on the narrow bit in the bottom far-left corner, you could see again that there’s a dose effect of seribantumab. Bigger doses cause more suppression. And then with afatinib, which is in the blue, that higher dose causes a completely flat line, so it’s still looking like afatinib is the one to watch. But of course, can you give those doses to a patient?

Alexander Spira, MD, PhD, FACP: Ross, I learned something when we were talking before. What do you think about the treatment time and days? How relevant that is? When you think about patients, 30 days is nothing. That’s what people are thinking. Do you want to comment on the differences in afatinib because you’re starting to see it rise around the 27-day mark? You’re starting to see the little peak. It’s delayed a little more in the seribantumab arm. Do you want comment on about that?

D. Ross Camidge, MD, PhD: When I look at these things, first, does it slow the tumor growth? If you get a big enough dose, do you get more of a flat line? Yes, you do. Do you even get some suggestion of the curve going down? Maybe you get some shrinkage—good. But then, what I called out the other day when we were chatting about this is, if you look at the curves that are more like a flat line, we’re not even at a month and the curves are starting to come up, it’s looking like acquired resistance. Not just delayed growth, but the growth dynamics have changed as if somehow it’s found a way around the treatment. The question is, is it finding a way around treatment more easily with seribantumab and that current dosing than it is with the afatinib? This raises the issue of what we’re going to see. Thirty days in a patient is nothing. Are we seeing the sped-up film looking at this mouse model? What I want to do is rebiopsy these tumors when they’re going up on the right-hand side and figuring out, what’s going on?

Let’s have a look. This looks a bit more like clinical data. This is almost that same model, but now they made it look like a waterfall plot. Rather clearly you can see afatinib causes shrinkage, but there’s a dose-response effect. Seribantumab is the same. There’s a dose-response effect, but it can cause shrinkage. This reassures me that if we were jumping in with a clinical trial, the response rate might be the appropriate end point. But that’s all I take away from this.

Alexander Spira, MD, PhD, FACP: I agree. When I first looked at these data, I had to be reminded about afatinib because there are a lot of drugs being studied in these NRG1 fusions. We’re not hearing much about afatinib anymore. There are some case reports and some anecdotal stories, but if you look at the previous slide—when we looked at the dose concentration—by the time you’re seeing tumor responses, that’s a pretty high dose. If you’re trying to come up with some results, a mouse is not a human and a PDX model is not a human. To me, given what we know about the talks of afatinib and the doses that you’re seeing here, that’s probably not a sustainable dose in real life. It explains to me why you have a better chance with seribantumab given the expected toxicities in the doses that you’re seeing. That’s what my take-home message is to look at this clinically.

D. Ross Camidge, MD, PhD: You’ve nailed it because everything so far looked like afatinib was the 1 to watch. But if you look at the mg/kg dose, we dose at afatinib at 30 mg, maybe 45 mg if we’re really gung ho. This is 15 mg/kg. There’s no way you’re going to be able to tolerate that dose in a human. Suddenly afatinib goes from the 1 to watch to this pipe dream that you can never achieve. That was the whistle stop tour through the preclinical data, and I hope you found that fun. Next, we’re going to put this all into context of the NRG1 infusions. Thanks, Alex, that was great fun.

Alexander Spira, MD, PhD, FACP: Thanks, Ross, I had a good time.

Transcript edited for clarity.