Guidelines and Approaches to HER2 Testing for Breast Cancer

Adam Brufsky, MD, PhD: Hello, and welcome to this OncLive® Peer Exchange®, “Expanding Treatment Options for HER2-Positive Breast Cancer.” I’m Dr Adam Brufsky, from UPMC Hillman Cancer Center in Pittsburgh, Pennsylvania. Joining me today in this virtual discussion are my colleagues Dr Carey Anders from the Duke Cancer Institute in Durham, North Carolina; Dr Virginia Kaklamani from the UT Health Science Center in San Antonio, Texas; Dr Rashmi Murthy from The University of Texas MD Anderson Cancer Center in Houston, Texas; and Dr Mark Pegram from Stanford Women’s Cancer Center in Stanford, California.

Today we’re going to highlight a number of topics pertaining to systemic treatment for HER2-positive breast cancer and the impact of recent clinical trial data on clinical decision-making. In addition, we’ll be covering key data from the ASCO [American Society of Clinical Oncology] 2020 Annual Meeting. Let’s get started on our first topic.

Is there still debate about HER2 [human epidermal growth factor receptor 2]? How do we test for it, or has that been settled? Mark, what do you think? Are we done? Do we have it, finally? Are the ASCO/CAP [College of American Pathologists] guidelines all, or are there any nuances we should think about?

Mark D. Pegram, MD: Sadly, the ASCO/CAP guidelines have been a work in progress. There have been multiple revisions, the most recent of which was in 2018. In that revision there were a few major changes. One is the definition for IHC2+, the equivocal IHC [immunohistochemistry] zone. They now say that HER2 testing on the surgical specimen is no longer mandatory if you’ve already done another core biopsy. There are more rigorous interpretation criteria, now for the gray zones, that is copy numbers of HER2 between 4 and 6 and ratios around 2. There is a much more defined workflow in the pathology lab to work those up. Finally, the expert panel for ASCO/CAP also recommended the use of dual-color FISH [fluorescence in situ hybridization] probes rather than ISH [in situ hybridization], which does not have a control probe for chromosome 17. Those are the major changes with ASCO/CAP 2018.

Adam Brufsky, MD, PhD: I’ll give you a perfect example, and I want to hear what you guys have to say. A woman comes in with a 4-cm tumor in her breast, and it’s ER [estrogen receptor] 5, PR [progesterone receptor] 0. It looks triple negative and the Ki is 50. We do a HER2 test on her, and it comes back at a ratio of 1.97 with a copy number of 4.2. What do we do? Virginia, what would you do in your institution if that came up? We get this every day, so that’s a question I’m asking.

Virginia Kaklamani, MD: We do, and I don’t know the answer. I always look at what the clinical trials have done and what patients have been included. This is hard because you have the central testing and you have the local laboratory testing, and results are different between the 2. Typically, I’d rather err on the side of giving anti-HER2 therapy than not giving it, but it’s really where we still need some data about what to do. It’s not black or white.

Adam Brufsky, MD, PhD: What do you think? Go ahead.

Mark D. Pegram, MD: When you are always on the fence like this, if you send the sample to an academic reference lab to really study the case in detail, oftentimes what you’ll find are examples of intratumoral heterogeneity. This was pointed out last year in the HER2 ASCO session in 2019. I was a discussant in that section, and it showed that in upward of 10% of the time there can be heterogeneity. You can have add mixtures of both truly HER2-positive and truly HER2-negative clones in the same primary tumor. It’s very important that this heterogeneity can both be geographic—so getting multiple samples of the same tumor is an advantage, in that respect—and also temporal. You can see that through selection pressure of HER2-targeted therapies there could be outgrowth of HER2-negative clones, for instance. That can be temporal. You’ll find that change in the metastasis compared with the primary. But at the time of diagnosis, unless you do multiple sampling and analyze it carefully, you may miss it.

Adam Brufsky, MD, PhD: What we do at our center to try to get around that is we have our guys count the most intense HER2-positive portion of the specimen. Rashmi, do you do that at MD Anderson? Do the guys do it there?

Rashmi K. Murthy, MD, MBE: We sometimes count more cells if we get an equivocal result like that. But I echo what’s been said before. I would probably verge on the side of treating or offering treatment in some of these equivocal cases just because of the impact that HER2-targeted therapy has had. The option sometimes is to get another biopsy sample to see if a different part of the tumor has a different result. That’s another option to rebiopsy and recheck the HER2.

Mark D. Pegram, MD: Another clinical pearl I like to call is the “If it walks like a duck” theory. Pathologic alteration of HER2 is clinically and statistically significantly associated with higher T stage, positive lymph nodes, high grade, decreased or absent steroid receptor expression, high Ki-67, and nonlobular histology. With the exception of pleomorphic lobular, it can rarely be HER2 amplified. If you have characteristics that fit all those or many of those criteria, and it’s a truly indeterminate gray-area case, you’re probably going to err on the side of treatment, as was mentioned earlier. If it doesn’t have any of these features, then I would say that probably.

Adam Brufsky, MD, PhD: Mark, I cannot agree with you more. For our audience, this is a really important point. That someone will come in and might have a HER2 of 1.8 and a ratio of 1.85, and you’ll treat them like a triple negative. There are people on this call who remember the days before we had trastuzumab. We would treat them, and nothing would happen. You’ve got these people coming with these incredibly horrible, very aggressive, what looks like triple-negative disease. And then you treat them, and nothing happens. You scratch your head, like, “What the hell is going on here?” And you put them on pembrolizumab and all this other stuff. In reality, you’ve got to be very careful. Carey, what do you guys think of this? That’s a really good point Mark is raising here.

Carey K. Anders, MD: The other piece of the ASCO Guidelines update in 2018 is if you have a grade 3, if it’s appearing like a very aggressive ER-negative tumor, to repeat the HER2 on the surgical specimen even if the core biopsy was negative. That gives you that second point in time to really ensure that we’re not missing the opportunity for this patient to benefit from HER2-directed therapy.

Adam Brufsky, MD, PhD: What happens if you have someone who comes in and is on the core of the HER2, again? It’s 2-plus copies, and the ratio is 1.8, 1.85. You do it—you give her a neoadjuvant chemotherapy—and nothing happens because you’re not giving her anti-HER2 therapy. What do you do then? You just give her adjuvant trastuzumab when you find out?

Carey K. Anders, MD: You could test the residual disease, right? You could test the residual disease and see if a HER2-positive clone emerges, absolutely.

Adam Brufsky, MD, PhD: That’s the issue. There is this heterogeneity issue. We really need to emphasize it to the people, and there was this whole concept that if it walks like a duck, it’s a duck. Those are important. Anything else that we missed in this part?

Mark D. Pegram, MD: Seeing is believing. You have your pathologist show you the data at tumor board. I can’t tell you how good that is. When you see immunostains that are 3-plus and not, seeing is believing. I’ve seen our own pathologists issue retractions to their original report, sometimes posted by a trainee, and the whole group gets to see it in the light of day.

Adam Brufsky, MD, PhD: I have tortured our tumor board for years on this very topic, especially when it gets to amplification. They say, “It’s negative.” And I say, “Guys, how many cells did you count?” I make them put it up. I make them show me what they did. Then they go, “Yeah, maybe we should count some more in this section.” Remember, our audience may not have the access to tumor boards that we all do.

Mark D. Pegram, MD: There’s nothing wrong with a second pathololgy opinion. You can send the sample to Mike Press’s lab at the Keck School of Medicine of USC in Los Angeles just as I do, and if there’s something really nuanced, he’ll call you on the telephone and explain it to you. I remember the first time he called me when there was an allelic loss of the chromosome 17 control probe. He said, “Be careful. The ratio is going to be high, but it’s not because of gene amplification. It’s the lesion of 1 of the control probes.” That’s superhelpful when you get that kind of feedback.

Transcript Edited for Clarity