Examining the HER2 Testing Puzzle: Fresh Research Explores Nagging Questions

In the more than 30 years since the human epidermal growth factor receptor 2 (HER2) gene was first successfully explored, HER2 status in patients with breast cancer has emerged as a durable and effective marker for evaluating tumors and selecting therapies.¹

The FDA has approved several methods of assessing whether the gene is promoting tumor growth, and HER2 gene amplification and/or protein overexpression has been variously estimated to be at work in 10% to 34% of invasive breast cancers, with the frequency associated with tumor grade and carcinoma status.^1-3

Established agents have proved effective in inhibiting HER2-positive forms of the disease, and now the treatment of such cancers is poised to enter a new era as strategies for dual inhibitors have emerged.

Yet key aspects of HER2 status in breast tumors remain a source of debate—and sometimes frustration– among researchers, despite the volumes of research that has been conducted and the hundreds of thousands of women worldwide⁴ who have been treated with HER2-targeted therapies, particularly trastuzumab (Herceptin).

Giuseppe Viale, MD, FRCPath, a professor in the Department of Pathology in the European Institute of Oncology and the University of Milan in Italy, wrote in a paper presented at the American Society of Clinical Oncology annual meeting in June 2011 that the oncology community should revisit some basic assumptions about HER2 status.⁵

He said outstanding issues include the accuracy and reproducibility of HER2 testing, scoring systems for HER2-positive disease, and the impact of tumor heterogeneity on HER2 status.

“Many oncologists seem to believe that pathologists are now infallible in assessing HER2 status of breast cancer, and that a very accurate identification of patients who are candidates for HER2-targeted therapies may be informed by the pathologic report,” he said. “Accordingly, oncologists are now already asking the next clinical question: whether it is possible to identify among the patients who are candidates those who will actually respond to the different anti- HER2 interventions.”

Viale, however, said tumors continue to behave in ways that “challenge our assumptions and dogmas.”

Edith A. Perez, MD

“This should prompt a renovated willingness to try and unveil peculiar aspects of HER2-positive breast cancer that have been apparently overlooked or misinterpreted until now,” he said.

Indeed, prominent researchers have been exploring HER2 controversies, with findings presented in recent months in journal papers and medical conferences that shed new light on testing issues.

The issues at stake are vital not only to patients with breast cancer but also to those with other tumor types, noted Edith A. Perez, MD, deputy director of the Mayo Clinic Cancer Center and a professor of Medicine at Mayo Medical School in Jacksonville, Florida.

“In this era of individualized medicine—the era of targeted therapies for patients with breast cancer and other malignancies—one of the most important aspects is to test for the target,” Perez said in an interview. “HER2 serves as an example of the management of patients today, and also the challenges that we will encounter in the future.”

Defining Criteria for HER2-Positivity

Perez has tackled one of the critical questions involved with HER2-targeted therapies: which criteria should be used to define the threshold of HER2 positivity.

“We don’t want to use therapies that have a low chance of working,” she said. “But I tell you it’s as important not to miss a single patient who may benefit from potentially lifesaving treatment.”

Breast cancer tissue assessed by fluorescence in situ hybridization, left, and immunohistochemistry, in samples tested at Clarient, Inc.

Photos courtesy of Clarient/GE Healthcare

Perez fears patients who could benefit from trastuzumab and other HER2-targeted therapies may not receive the drug because of confusion over what constitutes HER2-positive disease.

In December, Perez and colleagues published the results of a study in the Journal of the National Cancer Institute comparing the impact of the FDA criteria for defining HER2-positive disease and guidelines that the American Society of Clinical Oncology (ASCO) and the College of American Pathologists (CAP) developed jointly in 2007.⁶

Perez contends the ASCO/CAP guidelines resulted in a modification of the threshold the FDA endorsed in 1998 when trastuzumab initially was approved, and that now those guidelines are an underlying part of the laboratory standards for HER2 testing.

Investigators analyzed breast tumor samples from 2904 patients who had enrolled in the North Central Cancer Treatment Group N9831 study, a randomized phase III trial that began in May 2000 to evaluate trastuzumab as an adjuvant therapy for patients with HER2-positive resected early breast cancer.

The researchers found, among other things, that 3.7% of the patients identified as HER2-positive using immunohistochemistry (IHC) staining would not have been eligible for enrollment in the N9831 trial under the ASCO/CAP criteria, although they did qualify under the FDA definitions. (Table 1)

Table 1: Analysis of HER2 Testing Results

in Patients From N9831 Trial

FISH HER2/CEP17 ratio

HER2 IHC staining level

No. patients

(%)

<2.0

2.0 to

2.2†

>2.2±

Not done

2

221

74 (33.5)

16 (7.2)

125 (57.6)

6 (2.7)

3 and >10% to 30% strong membrane staining of tumor cells†

107

16 (14.9)

6 (5.6)

78 (72.9)

7 (6.5)

3 and >30% strong membrane staining of tumor cells±

2439

74 (3.0)

6 (0.2)

2287 (93.8)

72 (3.0)

^†FDA cutoff for HER2 positivity

^±Cutoff for HER2 positivity under joint American Society of Clinical Oncology and College of American Pathologists joint criteria.

FISH indicates fluorescence in situ hybridization; CEP17, chromosome 17; IHC, immunohistochemistry.

Adapted from Perez EA, Dueck AC, McCullough AE, et al. Predictability of adjuvant trastuzumab benefit in N9831 patients using the ASCO/CAP HER2-Positivity Criteria. J Natl Cancer Inst. 2012;104:159-162.

A false-negative rate of that magnitude potentially could affect 3000 to 5000 women annually in the United States, investigators said.

At the same time, however, the rate of false-negatives dropped to 1.0% if the patients deemed ineligible by IHC testing were retested using the fluorescence in situ hybridization method (FISH).

Overall, the study was limited by the small numbers of patients who were HER2-positive under FDA standards but not ASCO/CAP standards, and the impact of that difference in terms of disease-free survival could not be compared, the investigators noted.

Nevertheless, Perez believes there are two clear messages: that FDA criteria should be used to define thresholds for HER2-positivity, and that tissue should be retested reflexively if results on either test are negative or borderline.

“What we’re recommending is that physicians should consider something that is called expanded reflex testing,” said Perez.

Perez gave the example of a patient with a 3- cm, grade 2 tumor who tested negative for HER2 by FISH. “We recommend that those physicians should consider an expanded reflex testing, and by that I mean that in addition to FISH, they should do immunohistochemistry, just to be sure that this tumor is truly HER2-negative, to avoid missing the opportunity of identifying the HER2 positivity that may then lead to appropriate use of trastuzumab, which in turn may benefit the patient,” she said.

Indeed, the ASCO/CAP panel clarified its guidelines in 2011 in a Clinical Notice that recommended equivocal results on one type of test be tested using the other test.⁷ Likewise, the panel said it did not recommend withholding HER2-targeted therapy from patients who would have qualified under earlier generations of clinical trials.

Similarly, the National Comprehensive Cancer Center guidelines for HER2 testing incorporate checks and balances in testing borderline results.⁸(Figure 1)

Figure 1. NCCN Guidelines: Principles of HER2 Testing

IHC indicates immunohistochemistry; ISH, in situ hybridization

NCCN Clinical Practice Guidelines in Oncology Breast Cancer Version 1.2012. NCCN website. Published January 20, 2012. Accessed February 29, 2012.

Validating Initial Test Results

Underlying the debate about criteria for HER2 testing has been another key issue: the accuracy and reproducibility of laboratory results.

In 2005, researchers believed that up to approximately 20% of all HER2 testing worldwide yielded false-positives. The extent of false-negatives was not as clear but was estimated at 5% to 10%, according to Antonio C. Wolff, MD, an associate professor and research scientist at the Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins Hospital, Baltimore, Maryland.

Wolff, who co-chaired the ASCO/CAP panel that developed guidelines in 2007, said in an interview at the 10th International Congress on the Future of Breast Cancer in August 2011 that efforts to improve the fixation and handling of tissue specimens have resulted in dramatic improvements, at least in the United States. He said there are indications that the rate of false-positives in US test results has dropped to less than 6%.

Viale, however, expressed frustration with the state of HER2 testing in his ASCO presentation.⁵ “It is unacceptable that accuracy and interlaboratory reproducibility of the assessment of HER2 status continue to be worldwide concerns,” he said.

He said there remains a “very high rate” of interlaboratory discordance both with IHC, with as many as 15% to 20% of assays yielding false-positives, and with FISH.

Viale suggested that the rate of false-positives is minimized if “the tissue block selected for HER2 testing includes at least a minor component of non-neoplastic breast parenchyma.” More stringent compliance with guidelines and quality assurance standards, particularly in the formalin fixing of samples for IHC testing, also would improve accuracy, he said.

Figure 2. Potential Impact of Discordance in HER2-Negative Test Results*

^*Based on American Cancer Society estimates of 230,480 new cases of breast cancer in the United States in 2011 and 80% HER2-negative

Adapted from Vogel CL, Bloom K, Burris H, et al. Discordance between central and local laboratory HER2 testing from a large HER2-negative population in VIRGO, a metastatic breast cancer registry. Presented at: CTRC-AACR San Antonio Breast Cancer Symposium, December 6-10, 2011; San Antonio, TX. Abstract P1-07-02.

Researchers have studied the rate of false-positives by comparing the results of local versus central laboratory testing. The rate of false-negatives, however, has not been explored as fully, and recent research suggests cause for concern that patients are not being identified because of such readings.

In a poster presented at the CTRC-AACR San Antonio Breast Cancer Symposium in December, investigators reported findings from an analysis of test results in women who participated in VIRGO, an observational tissue substudy of 1265 patients with HER2-negative metastatic breast cancer.⁹

Of the 478 patients with HER2-negative tumors confirmed through central testing, 21 patients were found to have HER2-positive tumors, a discordance rate of 4.2%. When extrapolated across the number of new breast cancer cases diagnosed last year, that level of discordance would translate into 7744 patients in the United States and 46,487 patients worldwide who could miss treatment as a result of testing issues. (Figure 2)

The findings also indicated the importance of confirming results on one test with another test, said Melissa Brammer, MD, MPH, associate group medical director at BioOncology Medical Affairs, Genentech, and one of the study investigators.

Of the 21 patients whose test results were false-negatives, 17 had only one test, Brammer said in an interview. Of those 17 patients, 14 were positive upon retesting.

Melissa Brammer, MD, MPH

Correctly assessing the HER2 status of breast cancer patients becomes even more important as a new generation of therapies targeting the pathway takes shape, noted Brammer and others.

Genentech, which developed Herceptin, is seeking FDA approval for the use of pertuzumab as part of a dual inhibitor strategy in metastatic breast cancer, and is investigating trastuzumab emtansine (T-DM1), a novel antibody-drug conjugate. GlaxoSmithKline is studying lapatinib (Tykerb) in combination with trastuzumab in early-stage HER2-positive disease.

“We feel very strongly that patients are being potentially missed,” said Brammer. “With the nice results that we’re seeing with HER2-directed therapy, it’s important that you not miss these patients.

“A positive test doesn’t guarantee that you’re going to respond to HER2-directed therapy, but at least you have that opportunity to have the treatment,” she added. “If you test HER2-negative, you can’t get treated with HER2-directed therapy.”

Testing Technology Expands

Testing for HER2 status has been a subject of controversy ever since the FDA approved Herceptin in September 1998, along with a companion diagnostic, the Dako HercepTest, an IHC assay.

Although the question of whether IHC or FISH is a better test to assess HER2 status still surfaces, both tests are recognized in ASCO/CAP and NCCN guidelines. Wolff, the guideline panel co-chair, said the most important factor is for tissue samples to be handled and tested correctly by a laboratory that meets CAP accreditation standards.

Today, the FDA has approved more than a dozen testing systems, including kits and software, for assessing HER2 status, according to a search of its databases. Most of the tests use IHC or FISH methods; however, in recent months tests using new in situ hybridization (ISH) technologies have been approved.

These include the HER2 CISH pharmDx Kit from Dako, a dual-color chromogenic assay that uses a bright field microscope (as opposed to fluorescence microscopy), and the INFORM HER2 Dual ISH DNA Probe cocktail assay from Ventana Medical Systems, Inc, which detects HER2 and chromosome 17 on a single slide using a standard light microscope.

Brammer noted that IHC and ISH technologies measure slightly different facets of HER2 status. “HER2- positive breast cancer is a type of aggressive breast cancer that has on the surface of the cell these growth factor receptors. In HER2-phosphorous cancers you have an increased number of the receptors and so you can identify these receptors by immunohistochemistry [IHC], which actually is an antibody that identifies the HER2 growth factor protein on the surface of the cell.

“We believe that what leads to the production of too many receptors on the surface of the cell is an abnormality within the tumor cell called gene amplification; the chromosome 17 within the tumor cell is where the HER2 gene resides. And in HER2-positive breast cancer, where you have too much HER2 protein on the surface of the cell, it’s caused by having too many copies of that HER2 gene on chromosome 17.”

Genentech recommends repeat testing for negative results. “One or both tests can be used to determine whether someone is HER2-positive,” Brammer said. “But because they’re slightly different, we feel pretty strongly that if one test is negative, the patient’s tumor should be tested with the other test.”

Kenneth J. Bloom, MD

Heterogeneity Is Emerging Concern

No matter what technology is used, however, there are intricacies in HER2 testing that go beyond the results of any given test. In particular, researchers are focusing on tumor heterogeneity and the role it plays in assessing HER2 status during a treatment regimen.

Kenneth J. Bloom, MD, has been in the thick of research and debate about HER2 testing for more than a decade. Today, he is chief medical officer at Clarient, Inc, a leading cancer diagnostics company that offers both IHC and FISH assays for HER2. He participated in the VIRGO subset research presented at SABCS.

Bloom believes that the discordance between local and central laboratory testing that spurred the formation of the ASCO/CAP guidelines likely was affected by tumor heterogeneity. Although he believes standardization and appropriate laboratory protocols are important, he also feels there are nuances that must be taken into account.

“HER2 is the poster child of laboratory testing and the problems that you could get into with it,” he said in a recent interview. “I think today we recognize that there’s limitations in all tests and that it’s critically important not to miss somebody who is HER2-positive.

“What I’ve learned from the HER2 story is the most important thing to consider when interpreting a HER2 test result is the patient: How did they present? What’s their age? What was the differentiation of the tumor? What other factors are present? What’s the proliferation status of the tumor?” he explained.

“Everything has to be interpreted in context. If I get a positive result by any test, the patient’s eligible for therapy,” said Bloom. “What the clinical trials tell us is their likelihood of response is the same as anybody else’s who is eligible for therapy. So if they’re eligible even at that cutoff level, they’re eligible for therapy and should be treated.”

Moreover, Bloom said the cells selected for testing might affect the results because of tumor heterogeneity. “We now recognize that if a second block of tumor were assessed in a HER2-negative patient, that somewhere between 5% and 10% of the time, the second block of tumor is assessed as positive,” he said.

Changes in HER2 Status Studied

“

I think today we recognize that there are limitations in all tests and that it’s critically important not to miss somebody who is HER2-positive. ”

—Kenneth J. Bloom, MD

Even if a tumor sample has undergone reflex testing and therapy has been administered, the complexities of HER2 continue to be uncovered. In November, Niikura et al published findings that the research team called the largest data set thus far to examine whether patients’ HER2-positive status can change during treatment and what factors may affect such changes.³

Investigators analyzed 182 patients diagnosed with HER2-positive primary breast cancer between 1997 and 2008 at the University of Texas MD Anderson Cancer Center in Houston.

Upon review, 43 patients (24%) had HER2-negative metastatic tumors. The difference in status correlated significantly with whether the patients had received chemotherapy (P = .022) but not whether they had taken trastuzumab (P = .296).

The results confirm that patients whose primary tumors are HER2-positive upon initial testing can lose that positivity if the tumor metastasizes, the researchers noted. At the same time, they said the results might be affected by inaccurate testing, and that a large prospective study of HER2 status in metastatic tumors should be conducted.

Nevertheless, the researchers indicated, there is a clear message: “Our data strongly support the need for biopsies of metastatic lesions in primary HER2-positive breast cancer to accurately determine patient prognosis and appropriate use of targeted therapy.”

That sentiment already is taking hold in the oncology community. Brammer said NCCN guidelines now recommend that recurrent tumors be retested for HER2 status (particularly HER2 negativity) and for estrogen and progesterone receptor testing. “If that tumor is available for biopsy, it may change behavior and that can certainly guide treatment,” said Brammer.

At the Mayo Clinic, Perez and colleagues already have been implementing such strategies. “How much do we know about the changes that can occur between a primary tumor and a recurrent tumor? The preliminary data available tell us that even if we look only at estrogen, progesterone, and HER2 receptors, there can be a change in up to about 20% of the tumors from the primary to the metastatic,” said Perez.

As a result, she said, it is necessary to conduct biopsies of metastatic tumors for those markers, if possible, and store the tissue properly for future testing.

“One of the very important initiatives that we have at Mayo Clinic is to follow these data and recommend that if the patient develops recurrent breast cancer, that we retest the tumors,” said Perez. “In the next few years, we’re going to be beyond this. We’re going to do full genomic analysis of the 25,000 genes in the tumors to eventually figure out how to best manage patients.”

References

1. Ross JS, Slodkowska EA, Symmans WF, et al. The HER-2 receptor and breast cancer: ten years of targeted anti—HER-2 therapy and personalized medicine. The Oncologist. 2009;14:320-368.
2. Ross JS, Fletcher JA, Bloom KJ, et al. HER-2/neu testing in breast cancer. Am J Clin Pathol. 2003;120(suppl 1):S53-S71.
3. Niikura N, Liu J, Hayashi N, et al. Loss of human epidermal growth factor receptor 2 (HER2) expression in metastatic sites of HER2- overexpressing primary breast tumors. J Clin Oncol. 2012;30(6):593-594. Published ahead of print November 28, 2011.
4. Herceptin development timeline. Genentech website. http://tiny.cc/ZOAf2b. Accessed February 29, 2012.
5. Viale G. Controversies in testing for HER2. American Society of Clinical Oncology 2011 Educational Book. 47th Annual Meeting; June 3-7, 2011; Chicago, IL.
6. Perez EA, Dueck AC, McCullough AE, et al. Predictability of adjuvant trastuzumab benefit in N9831 patients using the ASCO/CAP HER2- Positivity Criteria. J Natl Cancer Inst. 2012;104:159-162.
7. ASCO Guidelines Clinical Notice. American Society of Clinical Oncology website. http://tiny.cc/VFAf2b. Published January 13, 2011. Accessed March 8, 2012.
8. NCCN Clinical Practice Guidelines in Oncology Breast Cancer Version 1.2012. Principles of HER2 testing. NCCN website. http://tiny.cc/rRAf2b. Published January 20, 2012. Accessed February 29, 2012.
9. Vogel CL, Bloom K, Burris H, et al. Discordance between central and local laboratory HER2 testing from a large HER2-negative population in VIRGO, a metastatic breast cancer registry. Presented at: CTRC-AACR San Antonio Breast Cancer Symposium, December 6-10, 2011; San Antonio, TX. Abstract P1-07-02.