Overview of HR2 Alternations in Solid Tumors 

We are going to turn our attention now to an overview of HER2 testing in solid tumors. Then I will follow immediately after with HER2 testing in GI cancers.

As I mentioned before, I am Rondell Graham. I am a GI/Liver and Molecular Pathologist at Mayo Clinic in Minnesota. I spent about 80% of my time doing GI/liver pathology and then the remainder doing molecular pathology focused on solid tumors and the majority of those being from the digestive system.

Brief Overview of HER2 Signaling

Here is a schema illustrating a brief overview of HER2 signaling. As many of you will know in the audience, it functions as a heterodimer on the surface of the epithelial cells. When there is binding of the appropriate ligand, you have this downstream signal triggering proliferation, survival, angiogenesis. Two of the commonly implicated pathways are shown toward the right side of the screen.

ErbB Family Receptor Tyrosine Kinases

Now the family of receptor tyrosine kinase, of which HER2 is a part, includes four members. You will recognize the name EGFR, which is the first ErbB2 or HER2 is the second. Then there is ErbB3 and ErbB4. These are typical receptor tyrosine kinase with an extracellular ligand binding region. Then there is a transmembrane-spanning region and then a cytoplasmic tyrosine kinase-containing domain.

The ligand binding is what initiates the homo or heterodimerization. Then you have this activation happening in the cytoplasmic kinase domain. There is some differences in how the signaling is transduced depending on the receptor, especially when you compare HER2 and HER3, HER3 having an impaired kinase activity and only activated upon dimerization with another receptor and HER2 binding any of the ligands and activated by heterodimerization with any other of the possible receptors.

You can see in the schema here the various proteins and a little bit of differences in their configuration and appearance.

What Is HER2-Positive NSCLC?

What is HER2-positive in non-small cell lung cancer. That is a question that is asked often in clinical practice because there is layers of complexity. I am delighted that Dr. Mackinnon is here to go through that in detail in his section.

At a high level, the considerations are, are we looking for patients where there is HER2 overexpression? That happens in about 15% to 30% of cases of non-small cell lung cancer. In that instance, you are looking for IHC expression. That is different from HER2 amplification, which occurs in a smaller proportion of non-small cell lung cancers. That is detected by a different methodology. So important to distinguish between overexpression and amplification.

Then the HER2 gene point mutations occurring in 1% to 3% of non-small cell lung cancers. So a very small proportion and detected again by a different methodology. It is important to have a clarity about which you are referring to when you are talking about HER2 positive. As I said, Dr. Mackinnon will go over this and bring clarity to this topic.

HER2 Testing in GI Cancers

Now we are going to turn our attention to the digestive cancers. That is my favorite area.

Case 1: Patient With Advanced Gastric Cancer

Here is a case from our clinical practice. The control is not included, but I would say that the control here was strongly positive. The positive internal control. The negative was negative. This is a sample from a patient with metastatic gastric cancer. This was the HER2 result that we obtained.

Poll 4

We have a polling question. This is one within the program. So not the pre-test. Which of the following most accurately describe your assessment of this sample and the next steps that follow? You found this in a lymph node. Go select from the answers A through D. What is your answer and what is the next step?

All right. Everyone got the answer correct. This is indeed a HER2 3+, so strongly diffuse expression. Essentially all the tumor cells are present in this particular metastasis. Indeed, this will support this patient being a candidate for HER2-targeted therapy.

Tumor Agnostic Prevalence of HER2 Alterations (Mutations and Amplifications) in GI Cancers

The prevalence of HER2 alterations, this includes both mutations and amplification in GI cancers, are illustrated here in this histogram. You can see that GE junction adenocarcinomas, second from the left, have one of the highest overall prevalences. You can see that the majority of these cases are altered by way of amplification, with a very small minority having multiple alterations and some having mutations.

Colorectal cancer, if you move a little ways to the right, that is a little bit about midway, approximately between 5% to 10% of cases will also have alterations with the majority showing mutations. Then as you move to new endocrine carcinomas, a majority of those, almost all of those are showing HER2 mutations. Then less frequently, the fourth box on the right, pancreatic cancer, also showing a low frequency of HER2 amplification and occasional cases of HER2 mutations.

It is important to recognize then that there is HER2 alterations across a broad spectrum of solid tumors and frequently for GI malignancies.

Guideline Recommendations for HER2 Testing in Advanced/Metastatic GI Cancers

In terms of the guidelines that are currently recommended for patients with GE junction adenocarcinoma, in this second column from the left, you can see that all patients, if advanced at diagnosis or metastatic disease, those patients are recommended for testing in all patients with colon cancer if they are presenting with advanced or metastatic disease and in all patients with metastatic pancreatic cancer, this testing is recommended.

For those patients not amenable to resection for the biliary tract cancers, this testing is recommended. You can see in each of those settings, IHC is recommended. But the additional tests that may be considered varies. NGS may be considered for GE junction adenocarcinoma. In our practice in Minnesota, we essentially do immunocytochemistry and ISH in all cases. We almost never do NGS in our clinical context.

For biliary tract malignancies, we will frequently do NGS and IHC, that is maybe less frequently. You can see how the distribution varies as we move across different types of pancreatic cancer because those specimens pre-resection or if they are not amenable to resection are typically very small. You can see that NGS is recommended more frequently.

HER2 Immunohistochemistry Scoring: ToGA Study

Here are the results from the ToGA trial which is a famous trial. Now I think many people recognize it. Here is a representative example of images of what a zero, 1+, 2+ and 3+ look like with the corresponding descriptions for a surgical specimen, that is a resection specimen and a biopsy specimen.

Because of the pre-test, we all recognize that if you have a tumor cell cluster with strong complete basolateral or lateral membranous staining, regardless of the percentage that will be considered positive in a biopsy in the context of GE junction malignancies, that will be considered a positive result or HER2 3+.

It is important to recognize that there is a distinction if you are dealing with a surgical specimen where you are dealing with a larger amount of tumor, in which case you are looking at greater than 10% of the tumor cells.

HER2 Testing: Colon and Other Sites

In other sites, this is actually a really interesting area. How we interpret these specimens. In the context of the colon and other digestive system malignancies, we are looking for 3+. There are a number of different scoring systems that have been proposed or used. In our particular practice, we actually use the GE junction scoring system to make our determination.

I would say we rarely, almost never will use FISH in our practice. We tend to focus on immunohistochemistry, and that has to do with what has been approved. One of the important things is to provide clarity what test you are doing and what your result is. If you are providing an IHC result, you want to describe whether you are seeing zero, 1+, 2+ or 3+ staining and if you are providing FISH results to provide clarity accordingly.

There are no specific cut-off values that have been determined for hepatobiliary tumors at present. Then there are various cut points used for tumors in different organ systems. There are different results for breast cancer and the different interpretation system than GE junction, as an example.

Now there is quite a bit of heterogeneity as you look across a tissue sample, and that is something that pathologists will have to grapple with in the interpretation of cases. But it is something that is feasible and can be overcome.

While most alterations will correspond, at least in the digestive system, to all the expression that you can visualize with immunohistochemistry, there are also patients who have activating mutations. We will hear probably more about that from Dr. Mackinnon speaking about lung cancer. That represents a layer of complexity that needs to be understood in providing results and interpretation and managing these patients.

Intratumoral Heterogeneity and Incomplete Membrane Staining of HER2

Here is a slide illustrated some of the heterogeneity and the extent to which it happens. If you can see here on this particular graph, if you look at the breast cancer study, you can see that the majority of patients had a homogenous 3+, whereas if you look at the results for gastric cancer, you can see there is great heterogeneity with different levels of staining see in many cases. That corresponds to my experience in clinical practice as well.

Tumor heterogeneity is much more common in gastric cancer than breast cancer. And incomplete membranous staining is more common, and of course, because the epithelium in the digestive system and in this instance, the stomach is quite a bit different from that seen in breast cancer.

HER2 Mutations vs HER2 Overexpression/Amplification in GI Cancers

Current approvals in gastric cancer and colorectal cancer are based on trial criteria requiring protein expression by immunohistochemistry. You can use immunohistochemistry, and FISH is not recommended. You do not want to bypass IHC and then go to FISH. You are going to start with immunohistochemistry, which fits into most routine anatomic pathology workflows really well.

The rate of HER2 mutations varies by tumor and location. There is a varying overlap with amplification. The presence of HER2 mutations may affect the response in patients who have those two alterations together, so both overexpression and mutation. There is some emerging data to suggest that this landscape is really complex, particularly in colorectal cancer.

CRC: A Molecularly Diverse Entity

Colon cancers are really molecularly diverse disease. This Venn diagram is particularly helpful. Maybe a majority of colon cancers have RAS mutations. Commonly they will be involving exons two to four. KRAS G12C is a minority of colon cancers but an important specific subset. Very rare cases have fusion genes including the NTRK fusion which is targetable. The majority of those cases that have NTRK fusions actually show dMMR or mismatch repair deficient MSI-high. That is a small but important group of colorectal cancers, some of which correspond to patients having this syndrome, some that are sporadic.

Of those sporadic dMMR cases, a majority of them have BRAF V600E mutations. BRAF V600E mutations, of course, can be found in patients with microsatellite stable. You can see them there in that purple or maybe magenta rectangle.

Then there are patients who are HER2 overexpressed or amplified. These patients are very commonly RAS wild-type. You see a small amount of overlap with some of the other illustrations you can see here in the diagram. Then the RAS wild-type patients. Again, a majority of the HER2 overexpressing amplified patients are RAS wild-type.

HER2 Is Overexpressed/Amplified in 2% to 4% of CRC

HER2 is overexpressed in about 2% to 4% of colorectal cancers. You can see a variety of different studies here of different sizes looking at different sites and a number of methods of detecting, showing a prevalence that range from of the order of 2% to 4%. So pretty consistently. Across various cohorts, you can see it is present on HER2 overexpression is seen in some patients who are RAS mutant, but the majority of patients with RAS wild-type.

This was detected also, too, in studies where we were not aware of the RAS status, but something that is consistently being seen in patients.

Let's Return to a Question From Earlier in the Program

We are going to return to a question from earlier in the program to see if we are able to transmit that information.

Posttest 1

If you have a non-surgical tissue biopsy, what are the criteria? What threshold will correspond to a patient who would receive or would be a candidate for HER2-targeted therapy? Your options are available here A through D.

Let us see how we did. All right. Substantial improvement with the correct answer showing that strong staining in the cell cluster regardless of percentage.

Posttest 1: Rationale

That is the correct answer as we illustrated in that slide from the ToGA trial. These slides will be available to you for review after the session. As we mentioned before, for our biopsies, so biopsy specimen, if you have a tumor cluster demonstrated strong complete membranous or basolateral or lateral staining, that would qualify as a 3+. Unlike surgical specimens, there is no minimum tumor cell percentage required.

With that, I am going to turn it over to Dr. Mackinnon.

HER2 Alterations and Testing in NSCLC

Dr. Alexander Craig Mackinnon (The University of Alabama at Birmingham Heersink School of Medicine): Thank you, and good morning, everyone. It is a pleasure to be here.

Case: Patient With Metastatic NSCLC

I am going to start with a patient that we had at UAB, is a 38-year-old woman. She presented with advanced high-stage adenocarcinoma of the lung. Her PD-L1 score was negative. We did molecular testing on her. The mutations that were detected and reported are listed here. The two ones we are going to focus on are the HER2 mutation. She had a missense mutation involving an amino acid that is in the extracellular domain of the molecule and HER2 copy number gains.

Poll 5

Based on these molecular results which you can see here on the slide, would this patient be eligible for an approved HER2-targeted therapy? You can key in your answer. Let us see what everyone thinks.

Great. There are therapies available for a patient like this.

What Is HER2-Altered NSCLC?

HER2 mutations or alterations in lung cancer come in two major types. The HER2 mutations, which are clinically the most significant, although from an incidence, the least common are present and also HER2 amplification overexpression. You see that about 8% to 23% of tumors.

Now talking specifically about the mutated HER2, there is some interesting clinical features of this. You see this more commonly in patients who have never smoked compared to other lung cancers that have driver mutations. These are more aggressive, and they have a higher incidence of brain metastases. Therapies against HER2-mutated lung cancer need to be able to reach the brain and treat tumors in that location.

HER2 Alterations: Breast Cancer vs NSCLC

This is a great opportunity to compare the mutation spectrum in lung cancer, which is a little bit different than we see in other tissues. Using breast cancer as an example, which is most common and most people are familiar with. You look for HER2 overexpression because that is a predictive biomarker for trastuzumab therapy. That can be done by IHC, where you look for protein expression, or through in-situ hybridization, where you look for gains in the number of copies of the HER2 gene.

In lung cancer, you can do both of those things too, but you also need to do, typically done by next-gen sequencing, some type of molecular assay where you can look for mutations in the HER2 gene. Something that is exciting and maybe not controversial, but uncertain about that is when you do next-gen sequencing, you do have the potential through the sequencing test to find out if the HER2 gene is amplified. It is unclear if that is as meaningful as the protein expression or FISH.

HER2 IHC Score Correlates With HER2 Amplification in NSCLC

There is a very strong correlation in lung cancer between the IHC score. As we heard earlier, we got a nice description of the gastric IHC score or breast cancer IHC. But cases that have 3+ IHC very often are also amplified, not always, but there is a strong correlation. Again, it is unclear where does next-gen sequencing, which can also measure copy number changes. How well does that correlate with IHC?

Distribution of HER2 Mutations in NSCLC

There is three major areas in the HER2 gene where we find mutations. The patient we started this part of the talk off with had an extracellular domain mutation. There is also mutations found in the transmembrane domain. Most importantly and clinically significant are mutations that are in intracellular portion of the molecule mainly clustering in the tyrosine kinase domain. These are the ones that are most clinically actionable.

There is drugs that work really well in patients with these mutations compared to the other sites. The mutations in the intracellular domain or the tyrosine kinase domain typically do not have other co-drivers in their cancers.

That bottom figure shows that the response, the different therapies that are listed on the left side of that graph worked the best in patients who have tyrosine kinase domains compared to mutations elsewhere in the molecule.

HER2 Testing Guidelines for Lung Cancer

Okay. The current NCCN guidelines recommend HER2 IHC expression and HER2 mutational analysis in all lung cancers. For patients who have gone on and resisted therapy, there is a recommendation that you can also do HER2 amplification to look for a potential resistance mechanism.

Guidance for HER2/ERBB2 Testing in Advanced NSCLC

Now HER2 is probably the latest member of a long growing list of important biomarkers in lung cancer, for which there is a targeted therapy. It is very important to have an assay that can work with FFPE tissue that can do all of these different mutations in one test. Oftentimes, especially with advanced disease, you get a really small biopsy. So you have to prioritize testing.

Knowing if there is actionable biomarkers is important because that is often prioritized ahead of using PD-L1 therapy. A fully informed clinical decision is made when you can have all this information and run a large panel that can detect variants in these different genes.

Targeting HER2 in NSCLC

It was really exciting in the last couple of years, new options for patients who have HER2-mutated lung cancer. Recently, and we are going to hear more about this with Dr. Dowell. There is some selective tyrosine kinase inhibitors, zongertinib recently was approved. These work best in patients who have tyrosine kinase domain mutations.

Then there is a second type of drugs or monoclonal antibodies. For example, in breast cancer, you would have trastuzumab. When those were initially tried in lung cancer, the response was not that great. Now there is a new class of drugs called antibody drug conjugates where you take those trastuzumab and you add on to a topoisomerase inhibitor. Those are showing great promise as well.

Beamion LUNG-1: Zongertinib for Advanced HER2-Mutated NSCLC

The trial that was recently done for zongertinib showed that this drug works best in patients who have a tyrosine kinase domain mutation versus, for example, our patient who had a mutation in HER2, but it was not in the tyrosine kinase domain. More recently, they were shown to work very similarly in patients who had been previously treated versus patients whose tumors had been untreated.

Now they just recently became an option for first-line therapy in patients with advanced metastatic non-squamous non-small cell lung cancer.

DESTINY-Lung01/02: Trastuzumab Deruxtecan in HER2-Mutated or HER2-Overexpressing NSCLC

The trastuzumab deruxtecan, which is an antibody drug conjugate, also shows activity in non-small cell lung cancer. To me, this is an intriguing paradigm where you have a drug that binds to the extracellular domain of the HER2 gene. However, it seems to work best in patients who have a tyrosine kinase mutation, which is intracellular. Some of the thought on how this could possibly be happening is that when the patient gets the drug, it causes the HER2 protein that is in the membrane to cluster. When that gets clustered, that causes the cell to start to endocytose these molecules and pull the drug into the cell.

Now, at this point, if the HER2 is wild-type, it is believed to get recycled back up to the membrane and take the drug back up with it. So it is actually leaving the cell and just go through this recycling mechanism.

However, when the HER2 is mutated, especially the exon 20 insertion mutations, then it is more likely to enter the lysosome, go into the endocytic pathway at which the linker between the drug and the antibody gets broken and the drug essentially delivers its payload to the cell, kills it, and then it can leak out and have a collateral effect and kill adjacent cells too. That is why it has been observed that patients with mutated HER2 respond better to the trastuzumab deruxtecan than patients who have high expression, although they do respond as well but just to a lesser extent.

DESTINY-PanTumor02: Trastuzumab Deruxtecan in Patients With HER2-Expressing Solid Tumors

Based on this, these drugs have been expanded to other diseases, including gynecological cancers and genitourinary cancers. The response rate is proportional to the amount of expression, whereas the HER2 3+ patients having the best responses.

Returning to the Case: Patient With Metastatic NSCLC

Going back to our patient, this top chart shows basically the molecular tumor burden. At the time she was diagnosed, that big blue balloon shows the level of mutations that were detected in her tumor. She was started on trastuzumab deruxtecan, went on that for several months and was started to have serial blood testing done, cell-free DNA.

She had a very nice response and went almost down to a complete molecular remission. At that point at our institution, we had a zongertinib trial that was started. Even though she did not have the exact tyrosine kinase domain mutation, she was still started on this drug and had about an 11-month good response where she had no detectable mutations in her blood. Then after about 11 months, she started to see the tumor coming back based on her molecular profile. At this point for the first time, a MET amplification was detected, which explains why she was becoming resistant to the drug. At the same time, there is now MET inhibitor, so this gives another option to potentially try a third-line therapy in this patient.

Applying HER2 Testing Results to Treatment Decisions for Patients With GI Cancers

I will now like to hand it over to Dr. Ku. Thank you.

Dr. Geoffrey Ku (Memorial Sloan Kettering Cancer Center): Thank you. That was super interesting. I learned a lot. For the next 10 minutes, this is a whirlwind tour of HER2 therapies in the GI cancer space.

HER2 in Gastric/GEJ Cancers

We will start with esophagogastric, which is the prototypical disease in this site.

ToGA: Chemotherapy ± Trastuzumab as First-line Therapy in HER2-Positive mGC

Many are familiar with this slide and these data. This was a ToGA study which was published 16 years ago now in 2010. It was a landmark study in gastric cancer for a number of reasons. It was the first study that demonstrated a benefit for adding a targeted therapy to chemotherapy. It was also the first study to show an improvement in median overall survival in the experimental arm beyond the psychological one year barrier.

This is a straightforward study. Patients whose tumors were HER2+ were randomized 1:1 to get trastuzumab, which we all recognize as the original anti-HER2 therapy, so chemotherapy with or without trastuzumab. You see the Kaplan Meier curve on the right which was in the intention-to-treat population.

Something that is especially relevant for pathologists is that this study actually used a different definition of HER2 positivity than the contemporary definition. At the time, any patients whose tumors were IHC 3+ or FISH-positive were eligible.

On the left, in the subgroup analysis, in a preplanned post-hoc exploratory analysis, we looked at outcomes based on patients with IHC 3+ and IHC 2+ FISH-positive or patients who were FISH-positive but at much lower IHC levels. You can see that the majority of the benefit were really in the patients who were IHC 3+ or 2+ FISH-positive.

How these data were treated at the time was that, in the US, trastuzumab actually was approved based on the eligibility population. In Europe, it was actually approved based on the subgroup showing the most benefit. Clearly with time, we now recognize that the patients who do benefit, that is the contemporary definition of HER2 positivity.

KEYNOTE-811: First-line Trastuzumab and Chemotherapy ± Pembrolizumab in HER2+ Gastric/GEJ

ToGA was in 2010 and it was not for another decade before we had another improvement in treatments. This was the KEYNOTE-811 study, which added pembrolizumab, so an anti-PD-1 antibody, to the combination of trastuzumab and chemotherapy.

The primary endpoints were progression-free as well as overall survival. An important secondary endpoint was response rate. I mentioned this because of the first interim analysis, there was a statistically significant improvement in response rates for the addition of pembrolizumab to trastuzumab and chemotherapy. This actually was sufficient for the FDA to approve the regimen back in 2021. This is actually the first time that a combination had been approved in oncology on the basis of improvements in response rates alone.

Subsequently, in the fullness of time, we do see that there are also improvements in progression-free as well as overall survival, but the approval actually has changed with time. The initial approval was in all patients irrespective of PD-L1 status. We now know with the final data that PD-L1, which is clearly an important predictive biomarker for patients receiving immunotherapy, is also predictive in this case.

In the Kaplan-Meier curves, the top row are all patients, and in the bottom row are patients whose PD-L1 CPS was one or more, and that comprised 85% of patients in the study, meaning that only about 15% of patients had tumors that were PD-L1 negative.

Now it is hard to see between the curves, but there is a slightly higher separation in the tumors that are PD-L1 positive. You can see that in terms of the hazard ratios. The hazard ratio in the top columns are 0.73 and 0.84, and these numbers go down a little bit correspondingly when you focus only on the PD-L1 positive tumors. The lower hazard ratio implies that there is more benefit in the PD-L1 positive tumors.

In fact, if you looked at the PD-L1 negative tumors, the hazard ratio for overall survival for the addition of pembrolizumab actually crossed one, although the confidence intervals overlap, which actually suggests that there may be harm to adding pembrolizumab to trastuzumab and chemotherapy in PD-L1 negative tumors. I certainly would not go so far, but I think it does reinforce that there is no benefit in tumors that are PD-L1 negative.

The approval of this regimen, therefore is conditional upon HER2 positivity but also PD-L1 positivity.

DESTINY-Gastric04: T-DXd vs Ramucirumab + Paclitaxel in HER2+ GC or GEJ Adenocarcinoma

Now moving from the first-line space to the second-line space, we have already heard about trastuzumab deruxtecan, and indeed I referred to it as the 800-pound gorilla, because it is now been evaluated in multiple HER2+ diseases and in multiple different scenarios.

DESTINY-Gastric04 is the culmination of several other studies that went before it. This was a randomized phase III study that was presented at ASCO last year. It evaluated T-DXd in the second-line setting and compared it to the then standard of care of paclitaxel with ramucirumab, which is an anti-angiogenic agent.

An important aspect of this study is that patients had to undergo a pre-treatment biopsy post first-line trastuzumab to confirm persistent HER2 positivity, as there are now multiple data sets that show that approximately one third, so 30% of tumors that are initially HER2+ can become HER2- at the time of progression. Therefore, documenting persistent HER2 positivity is critical.

DESTINY-Gastric04, simple study. Again, patients were randomized 1:1 between T-DXd and ramucirumab-paclitaxel, and there were clear improvements in outcomes with T-DXd. You can see that in the second-line setting median overall survival of 14.7 months, which is encouraging.

This is now firmly entrenched as the standard second-line regimen, although T-DXd had been approved several years prior to that on the basis of earlier studies.

HERIZON-GEA-01: Zanidatamab-Based Therapy in HER2+ mGEA

The other study that I want to focus on, which many may not have heard about, is a study called HERIZON-GEA-01. This is a study that was presented in January at GI ASCO and is in the process of becoming the new standard of care.

Zanidatamab, for those who have not heard of it, is a bispecific antibody. More specifically, it is what is called a biparatopic antibody, meaning that each arm binds to a different domain of a molecule, in this case, HER2. Conceptually, it is similar to having trastuzumab and pertuzumab, which is another anti-HER2 antibody. But instead of being on separate antibodies, it is on the same antibody.

Again, based on encouraging phase II data, this was a phase III study globally that evaluated zanidatamab. These are the three arms. Arm A was a standard arm of trastuzumab and chemotherapy, so the ToGA regimen. Arm B, the first experimental arm compared zanidatamab with chemotherapy. Arm C looked at zanidatamab plus tislelizumab, which is another anti-PD-1 antibody with chemotherapy.

The study had dual primary endpoints of progression-free as well as overall survival.

HERIZON-GEA-01: PFS

There is a somewhat complicated statistical hierarchical design which I would not have time to go through, but we will start by talking about the PFS data, both of which are positive is the bottom line.

If you look at the first comparison, which is arm B versus arm A, so it is a pure comparison of these two different anti-HER2 therapies. You can see that zanidatamab is clearly superior, and the median progression-free survival is improved by more than four months.

The other comparison, which is arm C versus arm A. Now we have added a PD-1 antibody. We actually see that numerically the improved median progression-free survival is exactly the same, 12.4 months as well. If you compare both studies, the hazard ratio is lower in arm C versus arm B, suggesting that immunotherapy is offering some benefits in terms of prolongation and benefit in terms of the tail at the end of the curve.

HERIZON-GEA-01: OS

In terms of overall survival. I will focus first on the Kaplan Meier curve on the right, which is the arm C versus arm A comparison. This is a statistically significant and clinically meaningful improvement in median overall survival now in excess of two years, 26.4 months, p value highly statistically significant.

Now when we make the arm B versus arm A comparison, there is also a numerical improvement. The hazard ratio is 0.80. The p value is not statistically significant. Now this is the first of several planned interim analyses. The next analysis is planned probably for the next several months. The expectation and hope is that with longer follow-up arm B versus arm A will also be statistically significant.

The one thing I would quickly mention is that in subgroup analyses, the interesting observation is that in terms of arm C, patients do seem to benefit irrespective of PD-L1 status. This regimen, either the doublet regimen or the triplet regimen is being evaluated by NCCN as well as FDA.

HER2 in BTC

HERIZON-BTC-01: Zanidatamab for Previously Treated HER2-Amplified Biliary Tract Cancer

Very quickly moving to other sites. In biliary tract cancer, there are also data for zanidatamab. The HERIZON-BTC-01 study was a single-arm study which evaluated zanidatamab in tumors that were HER2+. Again, we have already heard that in BTC, there is not a clear standard in terms of how to define HER2 positivity.

What seems pretty clear that is that in this case, all of the benefit was conferred to the tumors that were IHC 3+, which was about 80% of the tumors. In fact, if you look at the response rates, it is 52% for tumors that are 3+ versus 6% for tumors that are 2+. Similarly, progression-free survival is 7.2 versus 1.7 months.

On the basis of these data, the FDA approved zanidatamab in an accelerated fashion for tumors that are IHC 3+ and a phase III study in the first-line setting where it is added to chemotherapy is ongoing.

DESTINY-PanTumor02: T-DXd in HER2-Positive Solid Tumors (BTC Cohort)

The other anti-HER2 therapy that is also available in biliary tract cancer is T-DXd, and this is on the basis of a study called PanTumor, which evaluated multiple tumor types. Again, we also see here that all of the benefit seems to accrue to tumors that are IHC 3+ in terms of response rates, but also in terms of overall survival.

Therefore, T-DXd is also approved essentially for all solid tumors that are HER2+ IHC 3+.

Summary of Key Clinical Trials in HER2+ mCRC

Finally, in the time that I have left, this is just a quick summary of anti-HER2 therapies that are approved in colorectal cancer. Really the list is pertuzumab with trastuzumab, trastuzumab with tucatinib, which is another tyrosine kinase inhibitor as well as T-DXd. Now there are subtleties clearly to the choice of these regiments. For example, T-DXd is the one drug that is validated to have activity in the small subset of tumors that are both HER2+ as well as KRAS-mutated, whereas the others really have only been evaluated.

T-DXd also has data in patients who have received prior anti-HER2 therapy, whereas the other combinations are really only for patients who have not previously received anti-HER2 therapy.

There is also single-arm data in the first-line setting, suggesting that adding zanidatamab to chemotherapy is promising. This is not a regimen that is endorsed or approved at this time.

MOUNTAINEER: Tucatinib + Trastuzumab in Previously Treated HER2+ Metastatic Colorectal Cancer

Finally, this is the survival data from the MOUNTAINEER study which evaluated this combination of trastuzumab tucatinib. We see extremely high response rates of 40% in a treatment refractory setting, which you would never see with cytotoxic chemotherapy and a median overall survival, which is pretty remarkable at 24 months.

With that, I will turn things over to the final section on lung cancer. That is Dr. Dowell.

Let's Return to a Question From Earlier in the Program

Dr. Jonathan Dowell (Dallas VA Medical Center): There is one post-test question first. For patients with HER2 IHC 3+ metastatic biliary tract cancer that has progressed after first-line gemcitabine/cisplatin and immunotherapy, which of the following would be an FDA-approved option for targeted therapy? You see the choices there.

Posttest 3: Rationale

The correct answer is zanidatamab, as was highlighted by Dr. Ku. It is now FDA approved for biliary tract cancer that is previously treated.

Applying HER2 Testing Results to Treatment Decisions for Patients With NSCLC

Now we saved the best for last, so non-small cell lung cancer and how HER2 testing is pivotal in making our treatment decisions for those patients.

Biomarker Testing and Targeted Treatment in Advanced NSCLC

Many of you have seen a slide like this before, but just highlighting that we now know that a significant fraction of non-small cell lung cancer, especially non-squamous non-small cell lung cancer, will have a driver mutation or driver molecular alteration. Some of them are highlighted here. There are even a few that are not on this slide. We now have 25 or 30 approved targeted therapies in this setting. It is really critical that we know the molecular status of our patients, not even just in advanced stages, but increasingly in earlier stage disease as well.

Beamion LUNG-1: Zongertinib as First-line Treatment for Advanced HER2-Mutant NSCLC

I am going to highlight some of the therapies that are now approved for HER2-mutated disease in non-small cell lung cancer. This trial was talked about briefly earlier by Dr. Mackinnon, the Beamion LUNG-1 study, looked at zongertinib, a tyrosine kinase inhibitor as first-line treatment for patients with advanced HER2-mutant non-small cell lung cancer.

These were untreated patients importantly, so no prior therapy. They had metastatic disease. They all had to have a tyrosine kinase domain mutation in HER2. They were given zongertinib. Cohort two is the treatment naive group. The primary endpoints were overall response rate and then secondary endpoints you see there.

Beamion LUNG-1: Objective Response (Primary Endpoint)

This is the waterfall plot from that study. For those who may not look at a lot of waterfall plots in pathology, each line is an individual patient. The percentage shown is the decrease in their tumor size or increase in tumor size. As you see here, all the tumors essentially responded. So it was a pretty remarkable disease control rate of over 90%.

Those that are below the line at 30%, had what we consider a tumor response, and the response rate was exceptionally high, far higher than we see with chemotherapy in this setting. You see the responses appeared to be independent of whether they had the exon 20 YVMA insertion mutations or a different tyrosine kinase mutation.

Beamion LUNG-1: Zongertinib for Previously Treated Advanced HER2-Mutant NSCLC (Cohort 1)

This is the cohort that was previously treated from the same study. These folks had metastatic disease but had already had at least one line of therapy, and in many cases, multiple lines. Even in that setting, a very high response rate for this setting of over 50%, a disease control rate of 72%. The median PFS was quite good at 12 months.

Zongertinib Accelerated FDA Approval

Based on that, zongertinib was previously approved but recently got accelerated approval as first-line therapy for metastatic disease. This is the only agent that is approved in that setting currently for first-line treatment of metastatic non-small cell.

SOHO-01: Sevabertinib for Advanced HER2-Mutant NSCLC

I want to touch on the SOHO-01 study that looked at a similar drug, but a different tyrosine kinase inhibitor, sevabertinib, again an advanced HER2-mutant non-small cell lung cancer. This study also had multiple cohorts. Cohort D was previously treated but had not had HER2 therapy, so no HER2 targeted therapy.

Cohort E was previously treated with HER2-targeted agents, specifically trastuzumab deruxtecan in most cases. In cohort F, they were treatment naive. You see again very respectable response rates with this agent, a little less in cohort E where they had received prior HER2-targeted therapy and very reasonable progression-free survivals in this setting.

Sevabertinib Accelerated FDA Approval

On the basis of this study, sevabertinib also has received accelerated FDA approval. In this case, the patients have to have received a prior line of therapy, so it is not approved first-line.

DESTINY-Lung01: T-DXd in HER2-Mutated or HER2-Overexpressing NSCLC

I also wanted to touch on DESTINY-Lung01. The drug you have heard a lot about already, trastuzumab deruxtecan, an antibody drug conjugate, has also been evaluated in HER2-mutated or HER2-overexpressing non-small cell. In this study, these were folks who had advanced metastatic disease in either HER2 mutation or HER2 overexpression. They were refractory to all prior lines of therapy. This is a very heavily pre-treated group of patients.

Cohort one is the HER2 expressing group. Cohort two is the HER2-mutated group.

HER2-Mutated NSCLC Cohort in DESTINY-Lung01 Best Percentage Change in Tumor Size

Again, this is for the mutated group. You see again a very high response rate on this waterfall plot. Again, remember these are heavily pre-treated patients who typically do not respond to much of anything. The response rate was 55%.

In this study, they did allow extracellular domain mutations, the blue lines, and some of them responded as well. It was not just kinase domain mutations. A very respectable PFS of 8.2 months and median overall survival of 18 months, which is quite good in this heavily pre-treated cohort.

DESTINY-Lung01: Trastuzumab Deruxtecan in HER2-Overexpressing NSCLC

Smaller cohort in the overexpressing group and I would argue lower response rates. You see IHC 2+ or 3+ either orange or blue lines. The response rate was about 26%. Again, reasonable in a heavily pre-treated group of patients, but certainly not as high as we saw with the HER2-mutated group.

DESTINY-Lung02: Trastuzumab Deruxtecan in HER2-Mutated NSCLC

That study led then to DESTINY-Lung02, where they wanted to examine two different doses of trastuzumab deruxtecan in HER2-mutated disease. These were folks again with HER2-mutated disease. In this case, they had had one prior line of therapy, so just platinum-based chemotherapy and had progressed and they were randomized, as shown here, to either 5.4 or 6.4 milligrams per kilogram of trastuzumab deruxtecan.

The primary endpoint was overall response. You see here again very respectable response rates with both doses. But similar response rates with the lower dose, which is what ultimately ended up being approved. Across the bottom, I know it is small, but all of these folks essentially had exon 20 HER2 mutations or the vast majority did. It was restricted primarily to that population.

Again, you see there were a few folks that had extracellular domain mutations, but only a handful.

Trastuzumab Deruxtecan Tumor-Agnostic Approval

Then we have talked about this study previously that trastuzumab deruxtecan also now has tumor agnostic approval. In any patient who has HER2+ disease defined now by immunohistochemistry. We are talking about protein expression. If they are 3+, trastuzumab deruxtecan has activity in that setting. The approval was based on the combination of all three of these studies.

There was the DESTINY-PanTumor study that we saw earlier where there were very reasonable responses across a host of different tumor types. The DESTINY-Lung01 study that I showed previously, the HER2 3+ were the ones that tended to respond best. Then the DESTINY-CRC-01 study was in colorectal cancer. Again, you see a very reasonable response rate with HER2 3+ disease.

Trastuzumab Deruxtecan NSCLC Accelerated FDA Approvals

On the basis of this, trastuzumab deruxtecan has two approvals that affect non-small cell lung cancer. It is approved for patients with metastatic disease who have HER2 mutation and who have received a prior line of therapy. Again, not approved first-line, but approved as a subsequent line of therapy. It is also approved for patients who have HER2+ IHC 3+ solid tumors including non-small cell lung cancer, who have also received prior systemic therapy and do not have additional treatment options.

Let's Return to a Question From Earlier in the Program

Now we will return to a question from earlier in the program. Based on current FDA approvals, which of the following could be considered as first-line therapy for a patient with metastatic non-small cell and a HER2-activating mutation? Again, this is first-line therapy.

1. Sevabertinib;
2. Trastuzumab deruxtecan;
3. Zanidatamab;
4. Zongertinib.

Okay. Very good.

Posttest 2: Rationale

Yes. Zongertinib is the only agent of those that has first-line approval in this setting based on the Beamion LUNG-01 study that we reviewed earlier.

Poll 6

Then I was asked also to run through the remainder of these questions. Do you plan to make any changes in your clinical practice based on what you learned in today's program?

1. Yes;
2. No;
3. Uncertain.

Poll 7

Then please take a moment to text in one key change that you plan to make in your practice based on this education.

Q&A

Okay. I will turn it over to Dr. Graham to moderate if there are any questions.

Dr. Graham: Perfect. We have some time for questions, and we will just take them in the order in which they were received. I am grateful to the presenters for some really great presentations, and they have really stimulated some interest.

The first question is, what is the role of ctDNA biomarker testing to inform targeted therapy in GI and lung cancer? Dr. Ku, how does that inform your practice?

Dr. Ku: Sure. Focusing on GI cancers, the gold standard remains tumor NGS. ctDNA can be used as an adjunct in a number of ways. In rare instances, if you truly do not have access to biopsy or surgical specimen, then yes, ctDNA is something that we would fall back on.

It also can be used in certain scenarios. Dr. Mackinnon’s case was fascinating, in which, with ctDNA, you saw an increase in variant allele frequency and ultimately the emergence of a possible resistance mechanism. That is something that remains experimental in the sense that unless you clearly have well-defined mechanisms of resistance and you have well-defined therapies to overcome that, it is hard to know what to do with ctDNA data.

Certainly one scenario that occasionally comes up, where I am very reluctant to change treatment, is if someone is doing well and scans are stable, but something is going up on ctDNA, what do you do in that situation? I do not think we have good data. For that reason, it is not something I use in standard clinical practice, but as an adjunct or in very specific cases where there is no tumor tissue available, of course, it is a consideration.

Dr. Graham: Perfect. Anything Dr. Dowell?

Dr. Dowell: Sure. In lung cancer, we are frequently dealing with small biopsies. In my practice and many of my colleagues’ practice, we send ctDNA for molecular testing along with tissue. We send it simultaneously hoping that we will get a result from one, realizing the limitations as Dr. Ku mentioned of ctDNA. I use it quite frequently in that setting.

As was highlighted, where the field may be going as these assays improve is can we use that to monitor response and can we use it to trigger change in therapy, at least in lung cancer I do not think we are there yet and so I do not use it in that way. But that is likely coming at some point as the assays improve.

Dr. Graham: Perfect. Then the next question, Dr. Mackinnon. This seems tailor made for you. Can you get a HER2 mutation in NGS with a negative HER2 stain and what is the significance?

Dr. Mackinnon: Yes. There can be a discordance between NGS final mutation and IHC. This is, I will admit, a new area to me, but I believe from what we saw, Dr. Dowell can comment on this, that now the drugs can work in either instance. You can have the mutation or the IHC and they can both be indications for therapy.

Dr. Dowell: With trastuzumab deruxtecan we have that option if they are 3+ or mutation positive for the tyrosine kinase domain or the tyrosine kinase inhibitors. We do not have that data yet. It is critical to know the mutation status there.

Dr. Graham: Perfect. Next question. Do you perform IHC for both adenocarcinoma and squamous cell carcinoma of the lung routinely or only if requested? As a pathologist, for the adenocarcinoma of the lung, would you like it done routinely?

Dr. Dowell: I missed the question. Biopsy it at all or biopsy at progression?

Dr. Graham: I mean, do you want HER2 IHC perform routinely or only when you request it?

Dr. Dowell: Yes. In my practice, it is always great to have reflex testing, although we do not have that at our institution. We request what we want. Again, because I am concerned about the adequacy of tissue in many of my patients, I tend to request PD-L1 and NGS first and because I know the HER2 3+ and there is also an indication now for MET 3+ IHC. Those are subsequent lines of therapy. Then if I have enough tissue left over, I will send that later is what I typically do.

Dr. Graham: Okay, that is helpful for lung cancer. Do you suggest HER2 test for all colon cancers, same as MSI testing? Dr. Ku is not into the affirmative. Same thing is true in our practice. We will do HER2 on all colon cancers and MMR testing on all. Any variations?

Dr. Mackinnon: No, speaking to the lung cancer, it is new. I do not know that every lung cancer where I work is getting HER2 testing, but I think we are going to see that becoming more and more common.

Dr. Graham: Perfect.

Dr. Ku: I mean, everyone, with colon cancer gets next-generation sequencing. That is a one-stop shop. Hopefully, you will detect HER2 amplification if it is a HER2+ tumor, but also recognizing that HER2 positivity is much more common in the left side of the colon than right side, and so on and so forth.

Dr. Graham: Perfect. Do you ever attempt to score HER2-low in lung or GI cancers, and would this finding be actionable at this time?

Dr. Dowell: In lung, we do not have data yet for that. The answer is no that I do not use that clinically.

Dr. Ku: Yes, in GI, we have already seen with some of the data in colon and biliary tract that anything less than 3+, you typically do not get a great response. In gastric, I would say that there is currently interest in looking at this entity of HER2-low, in other words, IHC 1+ and FISH-positive, which would be considered HER2-low.

The only data come from actually from an initial study with trastuzumab deruxtecan. When you looked at the IHC 1+ FISH-negative, IHC 2+ FISH-negative patients. I mean, there was limited activity. It remains an area that is being looked at, but it is not the same dramatic activity that we have in so-called HER2-low breast cancer.

Dr. Graham: That is perfect. Yes. In our practice, we just report the results as they are from a pathologists perspective. Then the next question is, will you prioritize testing metastatic site biopsy rather than localized biopsy or resection specimens?

Dr. Mackinnon: That is a great question. What I was trained and taught is that the metastatic is probably the best one to test, because that might be the most up to date in terms of the mutation profile or it might be the most aggressive clones. But those are sometimes the least likely to be biopsied. It is a paradox in that sense.

Dr. Dowell: Yes. I mean, you end up balancing risk of a biopsy versus having tissue on hand from a prior specimen. If it is a really old surgical specimen that I am not sure is going to be successful in testing, then I will do a new biopsy and attempt to test that. The reality is that in lung cancer, again, we deal with the tissue that we have, whether it is the metastatic site or the primary site we will use what is adequate, even knowing that there could be heterogeneity in the molecular results between the metastatic site and the primary site.

Dr. Ku: The HER2 heterogeneity is well established. This is sometimes the argument for doing something like ctDNA testing, where, again, it really is a summation of all of the alterations. Sometimes you detect something that you do not find in whatever the primary biopsy is.

Dr. Graham: Perfect. We have time for two more, perhaps, the last two. Are there efforts to provide HER2 testing guidelines for lung or colon or biliary tract cancer, or are the guidelines for breast and gastric cancer sufficient? I am curious. They mean credit scoring criteria or test guidelines?

Dr. Mackinnon: I am not familiar of these. I read one paper where they compared the breast cancer and the gastric cancer and compared that, the IHC scores compared to amplification. For 3+, they are perfectly identical. The breast cancer IHC system teases apart 2+ and 1+ more than the gastric system does. So you start to see differences there. It can become somewhat arbitrary. It is interesting that in breast this ultra-low HER2 can have a response. It could just be tissue specific and maybe it just does not matter in other diseases.

It seems like in GI that the 3+ is really the most important. If it is not 3+, it is just very marginal benefit for those patients.

Dr. Graham: That is true. That is fair. The final question, how do you score and report HER2 on cytology specimens?

Dr. Mackinnon: I am going to plead the fifth on that one.

Dr. Graham: Yes. I am not aware of us doing that. We treat the cell block like a biopsy specimen and we will report it. Then if it is a breast cancer case, it will be reported with the breast criteria. All the other cases in general are reported with GE junction or gastric criteria.