The Unreliability of Zero Versus One Plus Results in Metastatic Disease

Dr. Antonio Wolff and Dr. Kim Allison discuss the latest ASCO-CAP guideline update on HER2 testing in breast cancer. This guideline update affirms previous recommendations, and provides commentary based on data from the DESTINY-Breast04 trial. Dr. Wolff and Dr. Allison review the questions from the oncology and pathology community raised by these results, and provide commentary on patients with HER2 IHC 1+ and 2+ and ISH-negative metastatic breast cancer. Read the guideline update, "Human Epidermal Growth Factor Receptor 2 Testing in Breast Cancer: ASCO-CAP Guideline Update" at www.asco.org/breast-cancer-guidelines.

TRANSCRIPT

This guideline, clinical tools, and resources are available at www.asco.org/breast-cancer-guidelines. Read the full text of the guideline and review authors' disclosures of potential conflicts of interest disclosures in the Journal of Clinical Oncology, https://ascopubs.org/doi/10.1200/JCO.22.02864

Brittany Harvey: Hello and welcome to the ASCO Guidelines Podcast, one of ASCO's podcasts delivering timely information to keep you up to date on the latest changes, challenges, and advances in oncology. You can find all the shows, including this one, at asco.org/podcasts.

My name is Brittany Harvey, and today I am interviewing Dr. Antonio Wolff from Johns Hopkins Sidney Kimmel Comprehensive Cancer Center and Dr. Kim Allison from Stanford University School of Medicine, co-chairs on 'Human Epidermal Growth Factor Receptor 2 Testing in Breast Cancer: ASCO-CAP Guideline Update'. Thank you for being here, Dr. Wolff and Dr. Allison.

Dr. Antonio Wolff: Thank you.

Dr. Kim Allison: Thanks for having us.

Brittany Harvey: Then, before we discuss this guideline, I'd like to note that ASCO takes great care in developing its guidelines and ensuring that the ASCO conflict of interest policy is followed for each guideline. The disclosures of potential conflicts of interest for the full guideline panel, including our guests on this episode today, are available online with the publication of the guideline and the Journal of Clinical Oncology, linked in the show notes.

So now jumping into the content, to start us off, Dr. Wolff, what prompted the update expert panel to revisit the 2018 ASCO-CAP recommendations on HER2 testing and breast cancer, and what is the scope of this update?

Dr. Antonio Wolff: Thank you, Brittany. We appreciate the opportunity of being with you today, and it's great to be here with my colleague, Dr. Kim Allison, as well.

What triggered this informatory update was the release from data in trial DESTINY-Breast04, which tested the antibody-drug conjugate trastuzumab deruxtecan in patients who in the past would have been considered to have HER2-negative disease. This ADC, trastuzumab deruxtecan, has a topoisomerase inhibitor payload that is linked to the antibody trastuzumab. And in the past, from all the previous data we had, trastuzumab alone, or in combination with chemotherapy, or as part of another antibody-drug conjugate T-DM1 was essentially active in patients with HER2-positive disease, which is traditionally defined as having overexpression of the HER2 protein, which almost by default is a result of gene amplification of the HER2 gene.

And what data from initial studies appear to suggest is that patients who would not be traditionally considered HER2-overexpressed or HER2-amplified were potentially benefiting or having evidence of clinical activity against this disease in the study of metastatic disease. And this was a randomized clinical trial for patients with metastatic breast cancer whose tumors were centrally determined to have IHC 1+ or IHC 2+ expression and would not have been called HER2-positive, would not have been called HER2-overexpressed. And for the tumors that were HER2 2+, they also had to have absence of gene amplification by an in-situ hybridization assay.

And what was very interesting is that there was a meaningful, clinically significant improvement in survival for that patient population. And that has some clinicians to begin asking whether there is a different subset of patients who would have in the past been called as having HER2-negative disease that now could potentially be a candidate for this drug. And is there a difference between these patients and patients who, in the past, would have been called as HER2-negative on the basis of IHC 0? And so what complicated things for us a little bit is that patients with IHC 0 were not eligible for this trial. And what is left unanswered by this clinical trial is whether all patients who don't have HER2 protein overexpression or HER2 gene amplification would potentially have benefited from this drug. Kim?

Dr. Kim Allison: Yeah, agree. I think the main impetus for the update was the exciting results from the DESTINY-Breast04 trial and the questions then that the pathology community and the oncology community had about whether this should change HER2 testing guidelines.

Brittany Harvey: I appreciate that background on what prompted the panel to revisit this guideline. So then, Dr. Allison, how did the panel come to the conclusion that the previous recommendations are current and valid?

Dr. Kim Allison: Right. So, as Antonio mentioned, the whole reason HER2 testing was first initiated was HER2-targeted therapies that showed response in the overexpressed or amplified tested population. And so guidelines have really been fine-tuned and crafted around distinguishing the HER2-positive for over-expression and amplification from negative. And this trial really questioned that in that maybe we can target lower levels of the HER2 protein, but this assay really wasn't designed for that. So we looked at the data from the trial and some of the limited other data that's out there and really came to the conclusion that, look, everyone in DESTINY-Breast04 benefited. The whole population benefited, whether you were 1+ or 2+. And because 0s were excluded from the trial, we don't know if they benefited.

So we thought it was premature to create a new category, a new result category, and change our current reporting to a HER2-low category, mainly because we don't know that there's a new predictive threshold for response to treatment. So essentially, what we've got is a trial that showed great benefit but didn't create a new biomarker that is predictive or prognostic. Instead, it repurposed this older test as a trial entry criteria. And so now we're kind of stuck with 1+ or 2+ ISH negative as their trial entry criteria that gives you eligibility for trastuzumab deruxtecan. So essentially, we reaffirmed our prior recommendations with acknowledgment that what these categories: positive, equivocal, and negative, refer to is for protein overexpression or gene amplification and that we should continue to use the same scoring criteria, 1+, 2+, 0, and interpretation as were used in DESTINY-Breast04 for their clinical trial criteria. But awareness is important.

Dr. Antonio Wolff: Yeah, the other thing that I would add, Brittany, I think we need to go back to what was the purpose of HER2 testing back in 1998 when we identified the survival benefit from trastuzumab in metastatic disease. And then, in 2005, when we had evidence of adjuvant benefit in improving disease-free and then overall survival for patients with early-stage disease. And the immunochemistry assays at that time were developed to differentiate between patients who had HER2 overexpressing disease or HER2 gene amplified disease versus not. At that moment, it was clearly identified that patients were considered as having HER2-positive disease that defined a biological entity, a tumor subtype, a group of patients who had worse prognoses in the absence of therapy. But then when they were treated with, for instance, chemotherapy with anthracyclines at that time, but then with HER2 targeted therapies with antibodies, these patients that otherwise would have a poor prognosis now were having an improved outcome.

HER2 was a marker of poor prognosis but also a marker of a good chance of deriving clinical benefit, so there was a predictive benefit. And everything else, IHC 0, 1 or 2+, ISH-negative, there has been no evidence that targeting the HER2 pathway was clinically important. Or even more meaningful, there was no evidence that these patients have– within the subset of patients that don't have HER2 positive or overexpressing disease - there has been no evidence that those patients have a different outcome based on low levels of expression of HER2.

A couple of years ago, in terms of trastuzumab, the NSABP reported findings in the data from NSABP B-47 where the antibody trastuzumab was added to chemotherapy to patients who were considered to have low levels of IHC expression, and in that case, was IHC 1+ or IHC 2+ gene non-amplified. And that settled the issue for sure, that there was absolutely no benefit in the adjuvant setting from the addition of trastuzumab. So we know that there's something different going on here. We know that if you now combine trastuzumab with a specific payload, in this case, the drug deruxtecan, which is a topoisomerase inhibitor, we are potentially targeting the HER2 protein. But these are tumors that are not considered HER2 addicted. These are not tumors whose biology is dependent on the HER2 pathway, so this is simply a better drug delivery.

And in this sense, data and evidence from Michael Press, a pathologist at USC that has done some seminal work with HER2 and HER2 testing, he once proposed that the vast majority of breast cancers have some level of HER2 present. And a lot of what is considered IHC 0 is an artifact related to suppression of the detection of the HER2 protein. So there's a chance that the tumors that are truly HER2 negative or HER2 0 are going to be a very small proportion. And IHC assays were never optimized to measure very low levels of HER2. They just don't have the dynamic range for that. And then, from a clinical standpoint, there is no evidence that different levels of HER2 when, in the absence of overexpression, identify groups of patients that have disease that have a different biologic behavior.

And I think, as Dr. Allison just mentioned, we don't have any evidence from the DESTINY-Breast04 trial that there is a differential benefit between IHC 1+ versus IHC 2+, ISH-negative. Those patients appear to equally benefit from therapy.

Brittany Harvey: Understood. That context is helpful in understanding this updated guideline.

So then you've both mentioned the category of patients who are HER2 IHC 1+ and 2+, and ISH-negative. So, Dr. Allison, this article includes a special commentary on those patients, those with HER2 IHC 1+ or 2+, and ISH-negative metastatic breast cancer. What are the essential points of this commentary?

Dr. Kim Allison: Yeah. So some of them we've brought up already that this test for HER2 IHC wasn't really designed to detect the low levels of protein expression that may be present in some breast cancers, including the all-important issue that the IHC 0s may not be truly null for HER2, that we may just not be sensitive enough to detect it, or there may be fixation and ischemia, time issues that are very subtle and create a false negative result, essentially by IHC that, and that in addition to not being necessarily predictive or prognostic, the 0 versus 1+ threshold, which really is a threshold that hasn't been tested yet.

But since eligibility for trastuzumab deruxtecan essentially now hinges around that 0 versus 1+ threshold, since these were clinical trial entry criteria, what can pathologists do to make best practice efforts to distinguish 0 from 1+? Because we felt like we should make some helpful recommendations, at least since this does appear to be a current status point that pathologists are going to be struggling with in practice. So the points we come up with in that commentary are to follow best practices, like making sure you're using the standardized ASCO-CAP Guidelines criteria. Making sure you pay attention to pre-analytic conditions of the tissue sample and that you're using controls with a range of protein expression, including 1+, to help ensure you've got an assay that's really looking at the right limit of detection since that has shifted somewhat in this instance.

And then for interpretation side, for pathologists to be sure to look on high power, so discriminating at 40x when you're trying to score a 0 versus 1+ stain since that's now relevant, you're going to need to go on high power and really distinguish from those two. And then, consider a second pathologist review in borderline or challenging cases or perhaps using additional tools. There's some additional tools online. There are learning sets that are out there to help with that distinction.

Dr. Antonio Wolff: What I would add to that is that I think, and Kim, I'm thinking of the pathologist now getting phone calls from oncologists saying, "Hey, Doctor Pathologist, you report this cancer as being IHC 0, and are you sure that this is truly IHC 0?" And I think we need to be careful not to put pathologists in an unfair situation. And I think we also need to be careful based on our behavior as oncologists that we could almost cause the extinction of the diagnosis of IHC 0 if pathologists feel somehow compelled to "try to help the oncologists" and potentially call a tumor that they would otherwise have called IHC 0, that they call that tumor IHC 1+.

And I think the reason for being cautious, as Kim mentioned, is these assays were not optimized for the ability to truly distinguish between IHC 0 and 1+. And we do not know if tumors that are IHC 0 clinically behave differently from tumors IHC 1+. Right now, that does not appear to be the case. And I think to a degree, we are being forced, based on the decision by the study sponsors to launch a study that excluded patients of IHC 0. We are left, I often say, twisting ourselves into pretzels, trying to come up with a way to discriminate between IHC 0 and 1+ simply because of the eligibility of the clinical trial and now the resulting FDA label for the study.

Because it is plausible that what if patients who had tumors that were IHC 0 had been included in this clinical trial? And what if we had determined that those patients also benefited from this new exciting antibody-drug conjugate? In that case, we would not be talking about creating new categories of HER2 low versus HER2 "ultra-low" and HER2 0, HER2 null. Because essentially, we would have identified a new clinical use for this exciting antibody-drug conjugate for patients who have tumors that are HER2 not overexpressed and HER2 not amplified. So, in fact, it is entirely plausible that the population of patients that could benefit from this antibody may go well above the original intent of DESTINY-Breast04. We just don't have the evidence at this point to say that those patients who would be called IHC 0 don't benefit. It's just that they were not included in the clinical trial.

Brittany Harvey: Well, your points there from both of you lead nicely into my next question, in that you've talked a little bit about how this impacts both oncologists and pathologists. So, Dr. Wolff, what does this guideline and commentary mean for both clinicians and patients with breast cancer?

Dr. Antonio Wolff: So I think what I would try to reassure oncologists, pathologists, and also patients and their caregivers and loved ones, over the last 20 years, I think we have seen a meaningful improvement in the quality of HER2 testing. And I think pathologists and oncologists recognize that breast cancer biomarkers, in general, in the past, were used purely for prognostic reasons or to complement anatomic pathology from a diagnostic standpoint. But now, many of these assays, especially ER and HER2, are used as the sole determinant of therapy selection in a binary fashion. If you are positive for ER, you can be a candidate for ER-targeted therapy. If you're positive for HER2, you may be a candidate for HER2-targeted therapy. And I think even though the current generation of IHCs were not equipped to make a differentiation between very low levels of HER2 expression from potentially no levels of HER2 expression, with all the limitations we just said, I think pathologists are today doing a very good job. They understand the importance of the work they do in helping us clinicians take care of patients in the clinic. As I often joke, pathologists are wearing the stethoscope with us. So I think we need to be kind to pathologists and not put them under the microscope, if you will, pun intended, or putting a lot of pressure on them. And I think I tend to trust the quality of the work they do.

I think there are two things that I would like to see happening. Number one, I would love to see the study sponsor allowing investigators to use a new generation of assays that are more quantitative to be able to back on DESTINY-Breast04 and test specimens of the patients that were triggered on the trial and see if there is a differential benefit in the observed outcomes of patients treated with trastuzumab deruxtecan according to levels of HER2, but that can be measured using a truly quantitative assay. And there are a lot of new assays out there. And I think the sponsors, they do have an obligation to all the patients who are participating in the trial to allow those things to happen. And the second piece is obviously to develop assays that are more quantitative than a traditional IHC.

And Kim, along these lines, a question that often comes up is what to do with patients who may have had a previous test that was IHC 0. And what should we, I guess, recommend to clinicians that they do with this situation?

Dr. Kim Allison: Yeah, I think this is a common question, and because of the unreliability of a 0 versus 1+ result, and we do see them change when you look at metastatic, or core versus primary surgical excision, 0 versus 1+ results shift around much more so than you'd expect if this was a true biologic difference. So I would look at a spectrum of samples across the metastatic progression. So if any of them are not 0, I think that's a result worth looking at. And considering that either there's heterogeneity there potentially or the 0s were false negatives and not consistent over time, so I would test the metastatic sample again. If you have a new sample, if that's 0, I would still consider treating based on a prior 1+ result or a different sample that's metastatic.

Dr. Antonio Wolff: Yeah, the one thing that I haven't done yet is actually for patients who have had– Let's say, that I have a primary term that was IHC 0, and then, they had, unfortunately, metastatic disease, and the diagnostic tissue that confirmed that they had metastatic disease also tested IHC 0. And now they are– unfortunately, disease has progressed after first or second-line therapy, be it anti-estrogen if they had ER-positive disease or chemotherapy if they had ER-negative disease. What I have not done is to request a new biopsy exclusively for the purpose of doing another HER2 test because in case the tumor had changed expression from 0 to 1+. Because what we don't know because of the variability, etc., Dr. Allison was just describing whether that change is real or not.

Again, it's really unfortunate that patients who were IHC 0 were not allowed for this study. There are other studies taking place right now looking at tumors that are more than IHC 0 and less than IHC 1+; that's DESTINY-Breast06. And those patients are being called by the study sponsor as "ultra-low". Although I am not a pathologist, but I have no idea how a pathologist can truly try using immunohistochemistry today; really reliably differentiate between it's not 0, it's not 1+, it's in between. I am just concerned that I think we may be asking or putting pathologists in a hard spot, asking them to do something with an assay that was not designed to perform that way.

Dr. Kim Allison: Even if we could get the interpretation perfect, to have digital tools to help us with interpretation, I think at that low level, IHC is just really sensitive to pre-analytics and analytic factors that are subtle. Even having a slide that's been sitting unstained for a week or so might change a 1+ to a 0 result. So it really is sensitive, maybe too sensitive, to those kinds of factors.

Brittany Harvey: Absolutely. Well, thank you both for those insights on what's facing clinicians and patients and the commonly asked questions today.

So then we've spent a lot of time talking about what's happened recently in this field. But, Dr. Allison, what are the ongoing developments and outstanding questions you're all facing regarding HER2 testing and breast cancer?

Dr. Kim Allison: Yeah. I mean, I think we've covered some of those. Is IHC 0 truly 0? Would it be responsive to T-DXd or other antibody-drug conjugates targeting HER2? And so if that is relevant, then there's a lot of work looking at maybe more sensitive or quantitative assays that are really designed to detect those lower levels of expression, unlike the current assays. And then digital image analysis to standardize interpretation if they are leading to differences. And then new standards to help us calibrate.

Brittany Harvey: Great. Well, I want to thank you both so much for all of your work to review and update this guideline on HER2 testing in breast cancer. And thank you for your time today, Dr. Wolff and Dr. Allison.

Dr. Kim Allison: Thanks for having us.

Dr. Antonio Wolff: Our pleasure. It's fun. Thank you.

Brittany Harvey: And thank you to all of our listeners for tuning in to the ASCO Guidelines podcast. To read the full guideline update, go to www.asco.org/breast-cancer-guidelines. You can also find many of our guidelines and interactive resources in the free ASCO Guidelines app available in the Apple App Store or the Google Play Store. If you have enjoyed what you've heard today, please rate and review the podcast and be sure to subscribe, so you never miss an episode.

The purpose of this podcast is to educate and to inform. This is not a substitute for professional medical care and is not intended for use in the diagnosis or treatment of individual conditions.

Guests on this podcast express their own opinions, experience, and conclusions. Guest statements on the podcast do not express the opinions of ASCO. The mention of any product, service, organization, activity, or therapy should not be construed as an ASCO endorsement.