Back Back
Diagnostic Gaps in HER2 Low Ultralow Breast Cancer
Bridging the Diagnostic Gaps in HER2-Low/-Ultralow Breast Cancer Pathology Reporting

Released: May 13, 2025

Expiration: November 12, 2026

Savitri Krishnamurthy
Savitri Krishnamurthy, MD
Liza Quintana
Liza Quintana, MD
Sara M. Tolaney
Sara M. Tolaney, MD, MPH
Activity Information

Activity

Progress
1
Course Completed
Continue
Key Takeaways
  • HER2-low/-ultralow scoring should follow the criteria laid out in the 2023 update of the ASCO/CAP guidelines.
  • Experts recommend reevaluating any tissue that is HER2-low/-ultralow if there is no staining in the metastatic tumor specimen to determine whether patients are eligible for treatment with trastuzumab deruxtecan.

The recent evolution of guidelines and recommendations surrounding HER2 classification and reporting in breast cancer has created diagnostic gaps, particularly with respect to HER2-low/-ultralow pathologies. Learn how leading experts are helping to bridge these gaps by reading their responses to questions during a recent live symposium.

What should one expect from a pathologist when testing?

Sara M. Tolaney, MD, MPH:
I believe that the DESTINY-Breast06–led change in reporting structure is helpful because pathologists are reporting HER2-ultralow (IHC 0 with incomplete membrane staining) and HER2-low (IHC1+ or 2+/ISH-) staining. However, a remaining challenge is that some specimens from patients with older biopsies were reported prior to the new system. Now healthcare professionals are asking, “Should this patient who has had every biopsy come back IHC 0 now be classified as HER2-ultralow?” The pathologist must be willing to evaluate the presence of HER2 staining from an older specimen. Some patients have heterogeneous disease. Because only 1 site is biopsied, imprecision is unavoidable.

If the results of the DESTINY-Breast15 trial (multicenter, global, interventional, open-label phase IIIb study of trastuzumab deruxtecan [T-DXd] in patients with HER2-low/-ultralow/-null unresectable or metastatic breast cancer) are positive, this could suggest that we do not need to distinguish degree of HER2 staining to select patients for T-DXd, although I wonder whether quantitative assays will be needed. Perhaps artificial intelligence (AI) technologies will allow a quantitative look at HER2 staining.

Savitri Krishnamurthy, MD:
When thinking about alternate ways of evaluating HER2 at the lower end of the spectrum, it is important to note HER2 RNA levels in tumor cells and HER2 protein levels, which can be assessed via mass spectrometry-based techniques, but neither should be considered standard of care, which remains IHC. All laboratories are encouraged to use the 4B5 antibody, and all the controls on the slide (not only 3+, 2+, and 0 but also 1+) should be used when staining.

Several AI-based decision support tools are available. Such tools can help maintain consistency and improve interobserver reading. However, although interobserver concordance may increase, the scoring is not perfect because humans trained the tools. There is no consensus on which AI tool is best, and not all laboratories have access to digital pathology. This is an evolving space. At present, we have to follow the criteria laid out by the 2023 update of the American Society of Clinical Oncology (ASCO)/College of American Pathologists (CAP) guidelines and report the findings following the CAP Breast Biomarker template updated in 2023.

What are the major pathology-related issues associated with HER2 testing in breast cancer?

Liza Quintana, MD:
HER2 test results can be challenging to interpret because of the slightly different thresholds used. However, there are websites that report HER2 scoring by experts that can be helpful by comparing your sample with theirs. Other issues involve analytical variables. If I have the opportunity to stain a metastatic sample as a cytopathologist, I prefer one from a surgical biopsy rather than a cytology specimen because HER2 expression can vary, and the biopsy has the advantage of being fixed in alcohol. 

Savitri Krishnamurthy, MD:
According to current ASCO/CAP guidelines, all patients diagnosed with metastatic breast cancer should have a biopsy taken at the site of metastasis and receive standard biomarker testing. In addition to the analytical variable noted by Dr Quintana, the antibody used could also affect IHC results. For example, the anti-HER2 4B5 antibody, which is now used as a companion diagnostic, was used in the DESTINY-Breast04 and DESTINY-Breast06 trials. Interpretation is another factor. The 2023 update of the CAP guidelines clearly stated there are certain best practices for identifying HER2-low/-ultralow breast cancers, some of which include evaluating the tissue at 40x magnification.

Three preanalytical factors should also be considered when testing: (1) the cold ischemia time, which should be less than 1 hour, (2) fixation, which should be performed in 10% neutral-buffered formalin, and (3) fixation time, which should be between 6-72 hours. Regarding bone biopsies, formic acid–based slow, decalcification, or ethylenediaminetetraacetic acid–based decalcification is preferred.

Pathologists should be careful to avoid pitfalls such as cytoplasmic staining with accentuation near the membrane, which is not membrane staining. Others include edge artifacts or staining near necrotic areas. Identifying true, incomplete membrane staining is key. When in doubt, it is good to ask your colleagues.

Has any variation in HER2 expression between the primary specimen and resection specimen been reported?

Savitri Krishnamurthy, MD:
There can be heterogeneity, especially at the lower end of the spectrum. Such heterogeneity is why there may be variation between the primary (core needle biopsy) specimen and excision specimen. Regarding HER2-low, some studies suggest that what is considered IHC 0 can be 1+ and vice versa. Regarding HER2-ultralow, I believe variation is expected because of the heterogeneity. This is why we recommend reevaluating any tissue that is HER2-low/-ultralow if there is no staining in the metastatic tumor specimen to determine whether patients are eligible for treatment.

How can you assess the impact of the bystander effect in terms of scoring? Does this relate to toxicity effects or drug activity? How can this information be used to optimize drug selection?

Sara M. Tolaney, MD, MPH:
I have seen patients with heterogeneous tumors where one area that has been biopsied exhibits HER2 staining and another does not. A response can be observed in both areas. Whether this is because the bystander effect or the fact that T-DXd still works in HER2-null tumors, I cannot say. One study looked at HER2 heterogeneity in the preoperative setting. Specifically, patients were given trastuzumab emtansine, which does not function via the bystander effect. Two sites in the primary breast tumor were biopsied. If there was heterogeneity, none of the patients achieved pathologic complete response, but if there was not heterogeneity, the pathologic complete response rate was approximately 60%. It was clear that heterogeneity affected response to preoperative therapy.

One could also retrospectively examine this with the DESTINY-Breast11 trial, which is examining preoperative T-DXd monotherapy or T-DXd followed by paclitaxel plus trastuzumab plus pertuzumab vs doxorubicin plus cyclophosphamide in patients with high-risk (T stage ≥3 and/or node positive or inflammatory) HER2-positive early breast cancer.

Savitri Krishnamurthy, MD:
There is another way to think about this effect. HER2-null or IHC 0 does not necessarily mean there is not a very low level of protein expression on tumor cells. The testing method may not be sensitive enough to capture tumor cells with very low levels of protein expression, meaning more sensitive methods may be required.

Tumor heterogeneity could be another reason why a tumor may respond even if there is no staining in any of the tumor cells.

If concurrent metastatic primary tumors have different HER2 results, does treatment follow the higher HER2 reading?

Savitri Krishnamurthy, MD:
In routine practice, the metastatic tumor is routinely tested for HER2 IHC. If the metastatic tumor is HER2 null, it is recommended to look at the results of archival HER2 testing that was performed on core biopsy or a resection specimen. A patient is eligible for treatment regardless of whether HER2-low/-ultralow is observed in the core needle biopsy, resection specimen, or metastatic tumor specimen. In DESTINY-Breast04 and DESTINY-Breast06, investigators evaluated the most recent biopsy, but when that was not available, they examined the archival core or archival resection tissue.

What do you recommend for reporting cytology basement in HER2-ultralow samples? 

Liza Quintana, MD
I report using the ASCO/CAP guidelines. If I have the option to select the specimen, I recommend the surgical pathology sample, knowing that with alcohol fixation and subsequent formalin fixation, HER2 results can be lower. When evaluating HER2-low/-ultralow, this slight difference in staining can make a big distinction.

The assay used may be able to be optimized for cytology samples. Dr Jeffrey Goldsmith published an article on the updated guideline regarding performing IHC on cytology specimens last year that I recommend referencing. If a cytology sample is all that is available, one could consider using another sample. It is definitely challenging.

Savitri Krishnamurthy, MD:
One of the main issues with cytology specimens is that the preanalytic factors are not as regulated as they are with surgical specimens. Therefore, in clinical trials, fine-needle aspiration specimens are not used. However, in routine practice, we do evaluate effusions. Fine-needle aspiration specimens are useful when a core needle biopsy specimen is not available. There may be some underestimation of HER2-low/-ultralow in cytology specimens, but we should all be following the same ASCO/CAP guidelines. There is much work to be done with respect to prestandardizing preanalytic variables with cytology specimens. 

Your Thoughts
What are your thoughts on the latest updates with HER2 scoring, and what are your biggest challenges with HER2 scoring today? Join the discussion by commenting below.

Continue

Poll

1.

Please rate your confidence in applying current ASCO/CAP guidelines and expert recommendations to identify patients with HER2-low-/ultralow breast cancer.

Submit