Listen to a discussion of this article through our AI-generated podcast.
CAP TODAY now offers the option to listen to a discussion of our articles with voices reminiscent of a Siri or Garmin speaker. In fact, the voices that you will hear are completely AI generated. Since this AI-generated podcast uses technology that is still in its early stages of development, there may be some errors in pronunciation and the tone may sound too upbeat for the material covered.
Karen Titus
September 2024—Every profession has its own news cycle. In pathology, the story arc is often a lengthy one, stretching from clinical triumph to Now what?
Such is the case with expanded use of trastuzumab deruxtecan, which was granted accelerated FDA approval in late April for adult patients with unresectable or metastatic HER2-positive (IHC 3+) solid tumors who received prior systemic treatment and have no alternative treatment options.
Three multicenter trials evaluated the efficacy of the drug: DESTINY-PanTumor02, DESTINY-Lung01, and DESTINY-CRC02. PanTumor02 looked at seven tumor cohorts: endometrial, cervical, ovarian, bladder, biliary tract, pancreatic, and other solid cancers (excluding breast, colorectal, gastric, and non-small cell lung cancers). The results of these trials often were striking (median duration of response for PanTumor02 was 19.4 months, for example), and physicians are now looking closely at this tumor-agnostic (also referred to as tissue-agnostic) indication.
Here was good news. “It’s really exciting, to be honest,” says Daniel King, MD, PhD, a medical oncologist at Northwell Health in New York. “It’s opening up an effective treatment for patients who previously didn’t have that as an option.”
Dr. King calls himself a big fan of the drug. His first encounters with targeting anti-HER2 came in the gastric cancer field, where it’s a well-recognized targetable biomarker. But agnostic tissue approval “is something that I’ve wanted to see for several years now,” says Dr. King, who is assistant professor of medicine at Northwell; adjunct professor of medicine, Cold Spring Harbor Laboratory; and director, research and development, Northwell Health Cancer Institute’s Center for Genomic Medicine.
And for his colleagues in pathology? That would be the next stage in the news cycle.
Says Jason Hornick, MD, PhD, professor of pathology, Harvard Medical School: “I would say the approval is still so new that I suspect most laboratories haven’t quite gotten their head around it.”
In TV terms, HER2 would be the Law & Order franchise of laboratory testing—around for decades and reinvigorated with new characters and spinoffs.
Breast cancer has long been the primary source of physicians’ experience with HER2-related therapies and testing. In the past decade or so, HER2 testing for gastric cancer and gastroesophageal adenocarcinomas has also become routine, or even universal, says Dr. Hornick, who is also director of surgical pathology and immunohistochemistry and chief of soft tissue and bone pathology, Brigham and Women’s Hospital, as well as the Mass General Brigham endowed chair in pathology. “But [for] other cancer types, it’s very new.”
That brings fresh challenges to labs, even given their longstanding familiarity with HER2 IHC in general. The fundamental question, says Dr. Hornick, is, absent a companion diagnostic in the pantumor setting, “How do you decide what’s a 3+ when you have so many different cancers and so many different assays?”
Granted, they’ll have time to figure out HER2 testing in the tumor-agnostic setting. As Dr. Hornick notes, “It’s new enough that we haven’t had many requests by my oncology colleagues at the Dana-Farber Cancer Institute.”
But when they do start rolling in, he says, “Even I’m not sure exactly what we’re going to do in terms of reporting.”
Dr. Hornick is comfortable with the antibody used in his laboratory, obviously, as well as with the results they report. “We’ve done a lot of cross-validation.” But, he adds, there are several different antibodies that have incredibly broad distribution among reference labs, commercial labs, academic labs, and private groups, all of which have different performance characteristics and results. “And because there’s not an FDA-approved assay that’s required for this new universal testing agnostic approval, you can use whatever assay you use, but then you have to decide how to report it.”
These issues, in Dr. Hornick’s view, weigh in as the biggest problems facing labs. “And they’re actually very big problems.”
Dr. Hornick says he’s puzzled by the lack of guidance for labs, as well as the failure to approve a companion diagnostic alongside the new indication for the drug.
“Even in the literature and in clinical practice for the cancer types that we’ve already been testing for HER2, for breast and upper GI, [where] we’ve had so much experience, we still have some confusion and debate about exactly what assay you’re using, what those thresholds should be,” Dr. Hornick says. “It takes practice, it takes training, it takes experience to have reproducibility among a pathology group.”
The problem with the pantumor approval, he says, will be especially apparent in large, high-volume groups. The subspecialists who will review cases in these additional tumor categories likely will lack experiences with HER2 IHC. “It’s going to be a steep learning curve, and it’s going to be hard.”
Dr. Hornick’s list of concerns is extensive. “How will we do the scoring?” he asks, given that there are different scoring systems for breast cancer, upper gastrointestinal adenocarcinomas, and colorectal adenocarcinomas. “All three of them look totally different in terms of how we assess the scoring for immunohistochemistry. So even though we have these approvals to talk about a positive result in terms of 3+, [and] everyone sort of knows what that means in general for HER2, there’s actually a different definition in the cancers that we’ve been testing for some time.
“So when you talk about tissue agnostic,” Dr. Hornick continues, “how do we decide what 3+ is? Are we going to use GI criteria? Colorectal criteria? Or something else?”
Given those challenges, “I don’t know why there’s not specific guidance,” Dr. Hornick says, though he anticipates pathology studies and consensus documents will emerge eventually. “And I suspect the CAP will probably engage experts to put together some type of practice guidelines, which they’ve done for other cancer types already for HER2.”
“It’s going to be a big problem,” Dr. Hornick reiterates. “And it’s not clear whom to ask for advice in that regard.”
In the meantime, he says, he’s trying to become that person for his colleagues. He and another senior colleague have been in deep discussion about tissue-agnostic testing, he says, and they planned to widen the discussion to the entire surgical pathology group at the next AP faculty meeting, “so we can serve as a resource for them if they get requests,” says Dr. Hornick, who is a member of the CAP Personalized Health Care Committee.
But, he acknowledges with a laugh, “I’m still going to have the same problem as my colleagues when they ask, ‘I got a request to do HER2 in the carcinoma type that I’m an expert in. I have no idea what to do about it—can you help me?’” They can run the test, he says, but anything short of a strong positive result will put pathologists in a quandary.
Arvind Rishi, MD, senior director, gastrointestinal and hepatobiliary pathology, and associate professor, Donald and Barbara Zucker School of Medicine at Hofstra/Northwell Health, Northwell Health Laboratories, says the tissue agnostic discussion came up in his laboratory early this year, when senior leadership and colleagues started developing a plan for how to report IHC, including what templates should be used.
Leadership wisely reached out to every service chief and director—in obstetrics-gynecology, GI, soft tissue pathology, breast, etc.—to ensure the reporting templates were standardized over every subspecialty system. “Obviously there were internal discussions and many questions,” Dr. Rishi says, and everyone recognized there would be challenges early on. Opening up clear communication between pathology and oncology teams was crucial. “We recognized that this is going to be an important patient management parameter that is going to be used more frequently.”
It was a smart move, and one every institution should go out of its way to support, says Dr. Rishi. “Have generous time for pathologists and oncologists to present these questions at tumor boards and at other forums. Create opportunities for them to develop more training data sets and CME around it. That sort of institutional encouragement is going to very much help in developing additional data to answer the questions around reproducibility and standardization. It will make patient care so much different.”
The challenges of tissue-agnostic testing are real, Dr. Rishi says. But he also sounds cautiously optimistic, at least for now.
With PanTumor02, “At this moment, and specifically speaking about Enhertu, it is a little bit easier,” Dr. Rishi says, given that there’s a sharp dividing line between cases that are 1+ and 2+, and 3+. “So I think it will be slightly easier to develop a standardized approach,” he says.
In addition to standardizing templates and guidelines, Dr. Rishi says, pathologists will be helped by having a low threshold for a consensus approach. If any pathologist thinks a case is a 3+, “then maybe it is shown to a more expert group that sees HER2/neu more frequently. You build a consensus and basically have more training on that,” he explains. “And titrate our observations as per the subspecialty sign-out setting.”
“Those are things we can do early on to have a relatively objective approach,” Dr. Rishi says. While any semiquantitative test can be challenging, “I think it’s going to get better very quickly, because as pathologists we are very used to immunohistochemical expression reporting.”
Using a two-tier system could be temporarily helpful, too, he says. “One thing I think is a good thing right now is because we don’t have much data on FISH and CISH in the 2+ category in solid tumors, now we at least have an opportunity to segregate the three-tier system and move to more of a two-tier system.”
“That’s a little bit easier route,” Dr. Rishi adds. But as experience with the 2+ category grows, “I think that question is going to come back again: Is 2+ then going to be reflexed to CISH or FISH?”
Even before the DESTINY trials, says Dr. Rishi, he and his colleagues had started doing HER2/neu IHC on metastatic colorectal carcinomas, at the request of their oncologists. It is now a reflex test. HER2/neu IHC is also performed on gastroesophageal adenocarcinomas. That gives him added confidence that with tissue-agnostic HER2 IHC testing, “The learning curve may be a little steep early on, but it will be really quick.”
What has he learned?
He points to three basic standardization steps. First, he says, “It’s important to mention what clone of HER2/neu immunohistochemistry we are using.” He and his colleagues use the 4B5 clone from Ventana. “We mention that in the report for standardization purposes.”
Every report also notes which tissue block has been used, he says. “It’s a good practice. So it can be compared or referred to as needed in the future for any further studies on molecular examination.”
A third standardization—not required, but useful, Dr. Rishi says—has to do with a preanalytical variable. “It should be a standard practice for every institution that the tissue should be in the formalin within 60 minutes of the procurement,” he says. This standard is derived from breast studies. “We don’t know that data on solid tumors in HER2. But it’s a general good practice to transcribe that experience from the breast studies, to keep it within 60 minutes cold ischemia time.”
Dr. King likewise is aware of the added burden tissue-agnostic testing will create for laboratories and outlines three main challenges his pathology colleagues are likely to face.
One: What would be the ordering process for all the locations within any given institution? Further, he asks, can that be standardized so that all providers know how to order the test when the need arises?
The second challenge, Dr. King continues, involves interobserver variability on the pathologic analysis: To what degree is it present, and how can it be minimized? “Especially when we’re talking about identifying lower levels of HER2 expression,” he says. “We need to have consistency in how doctors are doing the tests.”
His third concern has to do with tissue scarcity. “So now we have yet another demand on material. How do we make sure that we use the minimum amount, to allow for whatever tests are needed?” Exacerbating the issue, he says, is that the tissue-agnostic indication is for patients who’ve been previously treated, making it likely that requests for HER2 testing in this setting will come after much of the material has already undergone standard reflex and molecular tests.
There are, he says, “a lot of ways to chew on this.” He foresees a time when there will be an increasing number of molecular-defined or precision medicine approaches for cancer in general. “The tissue-agnostic approvals are going to have quite an impact,” Dr. King says, “because it will affect many if not all cancer types” as patients experience progressive illness. “But I think it also means an increasing demand on the baseline testing that’s performed for each new diagnosis.”
For resections, there may be sufficient tissue, Dr. King continues. Biopsies could be more problematic. “Proceduralists have already been given pressure to take as much material as they feel is safe” to meet the demands on tissue for molecular studies and some biomarkers. “I’m not sure how much more material we can get from an individual biopsy,” he concedes. Moreover, “Every pass of the needle can introduce some risk.”
One of his hopes, he says, is that oncologists will become more mindful of the demands these tests place on pathologists. One way to mitigate the gap is for both groups to discuss the best strategies for tissue use; clinicians would also benefit from more education on the topic. At Northwell, he says, pathology and oncology have worked closely to determine what types of reflex testing should be done. But missing from those chats, he says, is clinician awareness of how much material each test requires. “That’s something we could improve over time.”
Dr. Rishi shares a workflow he and his GI pathology colleagues use. When a needle biopsy is done, “the pathology assistants separate the fragments into two different paraffin blocks.” One is dedicated for molecular studies and the other for IHC. “That is something that has been working out very well,” Dr. Rishi says. “This is all part and parcel of good communication amongst the various teams internally within the pathology preanalytical scenarios.” This gets driven home during cross-training, he says. “When a pathologist assistant is training another pathologist assistant or student, to have a basic awareness that the tissue is a very, very precious thing. And if it’s multiple fragments of tumor tissue, then it needs to be segregated up front, instead of putting everything into one paraffin block.”
That approach will enable sufficient testing for both molecular studies as well as IHC. “Having a separate backup block is helpful for everyone—pathologists, oncologists, and, most importantly, patients,” Dr. Rishi says. “Because then we can prevent another procedure entirely.”
While observers marvel at the strong punch trastuzumab deruxtecan packs, they acknowledge the need to be more precise in determining where those punches land.
Even if the response rates are promising and the drug helps many patients, says Dr. Hornick, it’s also true that it will be ineffective for others. “If we can better define inclusion criteria and much better predict response, that would be wonderful.” Doing so will require much more benchtop and translational research, he says.
“This drug is not a panacea,” Dr. King says. Even in the higher levels of expression, for example, the results in the pancreas cohort of PanTumor02 were lackluster. Despite the accelerated approval of the drug for the tumor-agnostic setting, “Investigators and clinicians still need to be mindful that the responses we see differ by disease and by HER2 expression.”
The effectiveness in HER2-low settings is also likely going to be disease-dependent, Dr. King says. There’s promising activity in HER2-low gastric cancer, with HER2 low defined as IHC 1+ or 2+ with FISH negative. The DESTINY-Gastric01 trial included two exploratory cohorts for those two categories in late line gastric cancers. “It did show some activity, especially in the 2+ group, but not as much as we saw in higher levels of expression.” Nonetheless, wider use of the drug at lower levels is likely to follow.
In his own research, Dr. King is particularly interested in the use of trastuzumab deruxtecan in HER2-low expressing esophageal cancer, and he is involved in a clinical trial looking at this. More broadly, he says, “My interest is now that we have approval for HER2-positive pantumor, the next question is: To what extent can we, should we, investigate HER2-low tissue-agnostic approval?” In gastric cancer, for example, “there’s at least as many if not more patients who are lower-expressing HER2. One of the promises with trastuzumab deruxtecan is for it to work quite well in low-expressing tumors. So to me it would be exciting to try to extend DESTINY-PanTumor02 beyond HER2 plus to HER2 low.”
Before this drug, Dr. King says, “We’ve tried other antibodies outside of trastuzumab, including pertuzumab, zenocutuzumab, and margetuximab.” But T-DXd, perhaps due to its unique molecular features, “has opened up this HER2-low space in a way we didn’t know existed before.”
It’s early days for HER2 low, he and others acknowledge. But maybe one day, Dr. King says, “We’re going to be having another talk about approvals for the HER2-low space.”
As observers are making clear, the tumor-agnostic indication is a big advance.
The complications this advance creates are just as big—and it’s no surprise that labs find themselves trying to surf both waves, Dr. Hornick says. “Sadly, the evolution of cancer predictive biomarker testing in the U.S. has not been incredibly well-thought-through.” Pharmaceutical companies will often partner with a laboratory company on a clinical trial, then seek FDA approval for one assay without much input from the broader academic pathology leadership, he says, with their subject matter expertise and who can foresee the practical problems with multiple assays and multiple platforms. “For each of these biomarkers, we’ve gotten to the clinical trial approval without a lot of useful discussion among academic pathology and laboratory specialists.”
In fact, one potential project of the CAP Personalized Health Care Committee might be to establish an online resource for biomarker testing, “where we could try to bridge all these gaps,” Dr. Hornick says, “and provide as much information as possible for the assays, the antibodies, the drugs, the molecular testing.” The discussions are in the early stages, he says, but the hope is that such a resource would “help the pathology community get a handle on these really complicated issues.”
His hope is that the companies involved are homing in on the biological questions that have yet to be addressed. “These are incredibly promising results in a clinical trial,” he says. “But there are always additional, larger studies that often do fine-tune the indications.” And in the meantime, for these patients with no options left, “When you do an early clinical trial that has shown such striking results, you won’t want to withhold that from patients.”
Karen Titus is CAP TODAY contributing editor and co-managing editor.