Skip to main content
Download PDF

Accuracy of human epidermal growth factor receptor 2 (HER2) immunohistochemistry scoring by pathologists in breast cancer, including the HER2-low cutoff

HER2 IHC scoring concordance in breast cancer

Diagnostic Pathology volume 20, Article number: 35 (2025) Cite this article

Abstract

Background

Breast cancer was previously categorized as human epidermal growth factor receptor 2 (HER2)-positive (immunohistochemistry [IHC] 3+, IHC 2+ / in situ hybridization [ISH]–positive) or HER2-negative (IHC 0, IHC 1+, IHC 2+/ISH−). Recent studies of trastuzumab deruxtecan, a HER2-directed antibody-drug conjugate, have explored the spectrum of HER2 expression in tumors categorized as HER2-negative, including HER2-low (IHC 1+, IHC 2+/ISH−) and HER2-ultralow (IHC 0 with membrane staining). Clinical relevance of HER2-low and HER2-ultralow is reinforced by encouraging efficacy findings in these populations.

Objective

To assess HER2-low and HER2-ultralow scoring performance by pathologists, and compare real-world HER2-low scoring with centralized scoring by trained pathologists.

Methods

Formalin-fixed, paraffin-embedded breast cancer samples stained by the VENTANA anti-HER2/neu (4B5) Rabbit Monoclonal Primary Antibody (Roche) assay were selected to ensure adequate representation across all HER2 IHC scores (N = 500). Samples were rescored in a central laboratory by three pathologists trained in HER2-low scoring, and a majority consensus generated. Agreement between consensus and historical real-world HER2 scores was assessed by Fleiss’ kappa across HER2 scores (IHC 0, 1+, 2+, 3+).

Results

Substantial agreement was observed among central pathologists across HER2 scores (κ = 0.69), for the HER2-low cutoff (IHC 0 vs. IHC 1+, 2+, 3+; κ = 0.79), and the HER2-ultralow cutoff (IHC 0 absent membrane staining vs. IHC 0 with membrane staining, 1+, 2+, 3+; κ = 0.68). Substantial agreement was observed between real-world pathologists and central consensus for the HER2-low cutoff (κ = 0.72).

Conclusions

Pathologists can reproducibly score HER2-low and HER2-ultralow when supported by training. Findings may aid decision-making for patients with breast cancer who are potentially eligible for HER2-directed therapy.

Introduction

Following the availability of human epidermal growth factor receptor 2 (HER2)-directed therapies, patients presenting with metastatic breast cancer are routinely tested for HER2 expression [1]. Until recently, pathologists have classified breast tumors as HER2-positive (immunohistochemistry [IHC] 3+, IHC 2+ / in situ hybridization [ISH]–positive) or HER2-negative (IHC 0, IHC 1+, IHC 2+/ISH−) [2]. This classification is based on the efficacy of first-generation HER2-targeted therapeutics, which were approved for use in HER2-positive breast cancer [1, 3, 4].

Among tumors traditionally categorized as HER2-negative, there is a spectrum of HER2 expression [5]. In addition to tumors with no observable HER2 IHC membrane staining, the HER2-negative category encompasses IHC 1 + and IHC 2+/ISH − cancers [5] (commonly referred to as HER2-low) [6], as well as cancers with lower, but detectable, HER2 staining (i.e., faint or barely perceptible and incomplete membrane staining that is seen in ≤ 10% of tumor cells); these latter tumors are included in the category IHC 0 along with tumors that have no observable HER2 IHC staining according to the American Society of Clinical Oncology / College of American Pathologists (ASCO/CAP) guidelines [2]. Of note, the VENTANA anti-HER2/neu (4B5) Rabbit Monoclonal Primary Antibody (VENTANA HER2 [4B5]; Roche, Indianapolis, IN, USA) assay was optimized to detect overexpression of HER2 and not to detect basal levels of HER2 expression in normal breast tissue [7]. Therefore, a score of IHC 0 with membrane staining in breast tumor tissue indicates higher expression of HER2 than normal breast tissue, implying that HER2-directed therapies may be an effective treatment option.

HER2-low became of interest as a HER2 IHC scoring category relevant to patient care following the phase 1 J101 study [8] and the phase 2 DAISY trial [9]. These studies both demonstrated antitumor responses in patients with HER2-low advanced/metastatic breast cancer treated with trastuzumab deruxtecan (T-DXd), a HER2-directed antibody-drug conjugate [10, 11], suggesting that the antitumor activity of T-DXd might extend to the population of patients with HER2-low metastatic breast cancer.

The efficacy of T-DXd in patients with HER2-low metastatic breast cancer was confirmed in the DESTINY-Breast04 phase 3 trial [12], leading to the approval of T-DXd for adult patients with unresectable/metastatic HER2-low breast cancer who have received a prior chemotherapy in the metastatic setting or who have developed disease recurrence during or within 6 months of completing adjuvant chemotherapy in multiple countries and territories, including the United States, Japan, China, and Europe [13,14,15,16]. Additional HER2 cutoffs have been explored to determine whether patients with lower levels of HER2 expression are appropriate for HER2-directed treatment. T-DXd was shown to be an effective treatment option prior to chemotherapy in the DESTINY-Breast06 study, a global phase 3 randomized study that assessed T-DXd compared with investigator’s choice of chemotherapy in patients with hormone receptor–positive HER2-low metastatic breast cancer whose disease had progressed on endocrine therapy in the metastatic setting. DESTINY-Breast06 also included a HER2-ultralow subgroup with HER2 IHC membrane staining that is seen in ≤ 10% of tumor cells (IHC 0 with membrane staining), and results for this subgroup were consistent with the overall population [17]. Therefore, it is becoming increasingly important to understand if these new HER2 cutoffs (i.e., HER2-low and HER2-ultralow) can be scored reproducibly to ensure that patient treatment decisions are made in a manner that best supports clinical outcomes.

The VENTANA HER2 (4B5) assay was used in DESTINY-Breast04 and DESTINY-Breast06 to identify HER2-low and/or HER2-ultralow breast cancers and is validated for HER2-low and HER2-ultralow scoring. The VENTANA HER2 (4B5) assay is approved in several countries and territories as a companion diagnostic for T-DXd in HER2-low breast cancer, and is being evaluated for HER2-ultralow breast cancer [7, 18,19,20,21]. Overall agreement in HER2-low scoring between readers and between laboratories was 89.3% and 98.7%, respectively [7, 18]. Some published studies have suggested that HER2-low scoring may be more challenging than classic HER2-positive and HER2-negative cutoffs [22,23,24], whereas others have shown good reproducibility for HER2-low scoring, although training may further improve performance [25, 26]. Given that HER2-directed therapy is now available for patients with HER2-low breast cancer, it is important to verify that HER2 IHC scoring is accurate at the HER2-low cutoff.

Here, we conducted a real-world cohort study of HER2 scoring using a large sample set (500 clinical samples) across all HER2 IHC scores to assess reproducibility of HER2 scoring between pathologists at a central laboratory trained in HER2-low scoring across a range of cutoffs. Reproducibility of HER2-ultralow scoring between central pathologists was also explored. We then investigated how scoring by these central pathologists compared with historical scoring performed in routine clinical laboratories prior to HER2-low being defined as a clinically actionable patient category.

Methods

Generation of a reference standard

To create a set of reference samples to assess HER2 IHC scoring accuracy, samples were selected from a real-world cohort (N = 3750) of archival formalin-fixed, paraffin-embedded breast cancer tissue, acquired from biopsies, excisions, or resections. The cohort was made up of consecutive samples from patients with breast cancer tested between December 2018 and April 2020, stained using the VENTANA HER2 (4B5) assay following the manufacturer’s instructions for use and using BenchMark ULTRA instrument platforms [7]. Further information on the VENTANA HER2 (4B5) assay is provided in Supplemental Table 1.

All samples used in this study had previously been scored for HER2 expression, according to ASCO/CAP 2018 guidelines [2], by real-world pathologists and originated from one of three CAP / Clinical Laboratory Improvement Amendments (CLIA)-certified laboratories (samples acquired via Avaden BioSciences, Seattle, WA, USA). This scoring was performed in 2019 or 2020, prior to HER2-low being defined as a clinically actionable cutoff [27].

A subset of samples from the real-world cohort (N = 500) of 4B5-stained samples was selected to ensure adequate representation across all HER2 IHC scores (25% each at 0, 1+, 2+, 3+) and laboratories (one-third of samples from each), and rescored at a central laboratory (CellCarta, Antwerp, Belgium, a CAP/CLIA, ISO 15189 certified facility) using images of the previously stained 4B5 slides that were digitally scanned at the local sites at ×40 magnification (Leica Biosystems Aperio Scanners). Three central pathologists rescored the resulting images after receiving specific training (details below) in the scoring of HER2-low per the ASCO/CAP 2018 guidelines [2]. The three central pathologists included a board-certified gynecological cytologist with > 20 years’ experience, and two surgical pathologists with a background in breast cancer. The scores were then used to generate a reference standard HER2 IHC score (0 absent membrane staining, 0 with membrane staining, 1+, 2+, or 3+) for each sample by consensus, with agreement of at least two out of three pathologists required to generate a consensus score. The central rescoring generated a new cutoff in the dataset (IHC 0 absent membrane staining vs. IHC 0 with membrane staining, 1+, 2+, or 3+) that was not present in the historical dataset.

Training received by central pathologists

Virtual web-based training in scoring of HER2 per ASCO/CAP 2018 guidelines [2] was provided by CellCarta. The three central pathologists received a 1-hour theoretical overview of the ASCO/CAP 2018 HER2 scoring algorithm, with specific attention on the IHC 0, 1+, and 2 + definitions and focus on common scoring difficulties. The presentation also introduced the concepts and definitions of HER2-low (IHC 1 + and IHC 2+/ISH−) and HER2-ultralow (IHC 0 with membrane staining). Following the overview session, pathologists received access to a web-based PathoTrainer platform (Pathomation, Berchem, Belgium), through which they conducted virtual training and assessment. Pathologists first reviewed five pretest cases with the HER2 IHC consensus score provided, and then scored 30 test cases according to ASCO/CAP 2018 guidelines, adapted to differentiate IHC 0 into IHC 0 absent membrane staining and IHC 0 with membrane staining. Each case consisted of three images: a hematoxylin and eosin slide, a HER2 IHC slide, and a negative control slide (i.e., a slide stained in parallel with the HER2 IHC slide but using negative control reagents). An 85% concordance with the consensus score was required to pass the training. If the 85% concordance threshold was not met during assessment, pathologists received a one-on-one discrepant case review session, which had to be completed before the training was passed.

Scoring agreement between central pathologists

To assess HER2 scoring agreement between the three pathologists at the central laboratory, Fleiss’ kappa was calculated to assess the relative strength of agreement (i.e., κ > 0.8, almost perfect; 0.6 < κ ≤ 0.8, substantial; 0.4 < κ ≤ 0.6, moderate; 0.2 < κ ≤ 0.4, fair; 0 ≤ κ ≤ 0.2, slight; κ < 0, poor) [26, 28]. Overall percent agreement (OPA) and 95% confidence intervals (CIs) for OPA were calculated using standard methods to assess percent agreement between central pathologists.

Historical scoring accuracy

To assess HER2-low scoring performance of real-world pathologists, the historical HER2 IHC scores were compared with the reference consensus scores for the centrally rescored cohort samples. The historical HER2 IHC scores had no separation of IHC 0 absent membrane staining compared with IHC 0 with membrane staining; to allow for comparison, the same criteria were applied to the central consensus scores, and both were grouped as IHC 0. Percent agreement between real-world and consensus scores was determined by calculating OPA, negative percent agreement (NPA), and positive percent agreement (PPA) using standard methods, with central consensus score as the reference score.

To account for potential confounding caused by the equal distribution of scores in the subset of 500 samples, the real-world HER2 IHC score distribution in the full set of 3750 samples was determined, and a corresponding weighting applied to calculations performed on the 500-sample cohort (Supplemental Table 2). The calculated distributions for each HER2 IHC score were used as linear weighting factors in the OPA/NPA/PPA calculations (considering observed/expected frequency of each HER2 IHC score).

Impact of test laboratory, sample type, and tumor type on scoring agreement

The impact of test laboratory, sample collection method, and sample tumor type on HER2 scoring agreement was assessed by calculating Fleiss’ kappa for each subgroup.

Results

Inter-reader agreement in HER2 scoring for central pathologists

Scoring agreement was assessed among three central pathologists trained for HER2-low scoring of breast cancer tumor samples. Information on tumor stage, sample location, and sampling procedure for the cohort rescored to generate the reference standard is provided in Supplemental Table 3.

A consensus was reached for 484 of 500 breast cancer samples (including biopsy and excision samples) by majority voting based on the scoring results of all three pathologists (Table 1). Ten samples (2%) were deemed to be non-evaluable by all or the majority of pathologists, and for six samples (1.2%), a consensus score could not be reached (Supplemental Table 4). In all six cases, at least one pathologist marked the sample as failed due to no/insufficient tumor or poor image quality. It was decided that these 16 non-evaluable/non-consensus samples should be excluded from the subsequent analysis.

Table 1 Central pathologist agreement across HER2 IHC scores for breast cancer samples, used to establish consensus
Full size table

Substantial agreement among central pathologists trained in HER2-low scoring was observed across HER2 scores (IHC 0, 1+, 2+, 3+; κ = 0.69) as well as for the HER2-low cutoff (IHC 0 vs. IHC 1+, 2+, 3+; κ = 0.79), with OPA ranging from 85 to 97% for all scores, and from 91 to 94% for the HER2-low cutoff (Table 2). For the novel HER2-ultralow cutoff (IHC 0 absent membrane staining vs. IHC 0 with membrane staining, 1+, 2+, 3+) explored in DESTINY-Breast06 [17], substantial agreement was observed (κ = 0.68), and OPA ranged from 96 to 97% (Table 2).

Table 2 Inter-reader agreement for central pathologists with HER2-low training
Full size table

Agreement between historical HER2 real-world scoring and central consensus score

Moderate agreement (κ = 0.60) was observed between historical scoring by real-world pathologists and the central consensus score across HER2 scores (IHC 0, 1+, 2+, 3+); substantial agreement (κ = 0.72) was observed for the HER2-low cutoff (IHC 0 vs. IHC 1+, 2+, 3+). OPA (95% CI) was 90% (87–92%) for the HER2-low cutoff (Fig. 1), and 86% (85–87%) when weighted to reflect the real-world distribution for each HER2 IHC score (Supplemental Table 5).

Fig. 1
figure 1

Agreement between historical real-world HER2 scoring and central consensus score. a NPA and PPA were defined using central consensus score as the reference score. CI: confidence interval; HER2: human epidermal growth factor receptor 2; IHC: immunohistochemistry; NPA: negative percent agreement; OPA: overall percent agreement; PPA: positive percent agreement

Full size image

Impact of test laboratory and sample type/location on HER2-low scoring agreement

Agreement with central consensus varied between local testing laboratories (κ = 0.59–0.81), by surgery type (biopsy vs. excision), and tumor type (primary vs. metastatic) (Fig. 2). Substantial agreement was seen using core-needle biopsies or other biopsies (κ = 0.75). Substantial agreement was seen with metastatic and primary samples (κ = 0.95 and κ = 0.68, respectively), although the sample number for metastatic tissue was too small to draw strong conclusions. Lower agreement was observed using samples obtained by excisions or resections (κ = 0.52), although sample numbers were too small to draw strong conclusions.

Fig. 2
figure 2

Agreement between local and consensus HER2 scores. Scores are broken down by test laboratory, surgery type, and tumor type for all evaluable samples. CNB: core-needle biopsy; HER2: human epidermal growth factor receptor 2

Full size image

Discussion

In this study, we found substantial agreement across HER2 scores between central pathologists trained for HER2-low scoring (IHC 0, 1+, 2+, 3+; κ = 0.69), including for the HER2-low (defined herein as IHC 0 vs. IHC 1+, 2+, 3+; κ = 0.79) and HER2-ultralow (defined herein as IHC 0 absent membrane staining vs. IHC 0 with membrane staining, 1+, 2+, 3+; κ = 0.68) cutoffs. Inter-reader agreement at the HER2-ultralow cutoff measured by OPA was comparable with or better than that at the HER2-low cutoff. This high level of agreement increases confidence that the HER2-low and HER2-ultralow cutoffs can be reproducibly scored for clinical decision-making with appropriate training. This is important to ensure that the right patients are being selected for treatment following the approval of the HER2-directed antibody-drug conjugate T-DXd for the treatment of HER2-low unresectable/metastatic breast cancer and the DESTINY-Breast06 trial that included patients with HER2-ultralow unresectable/metastatic breast cancer.

The investigation and subsequent approval of T-DXd for treatment of HER2-low metastatic/unresectable breast cancer followed observations from the J101, DAISY, and DESTINY-Breast04 trials. In the phase 1 J101 study, antitumor responses to T-DXd were observed in 20 of 54 patients (37.0%) with HER2-low advanced/metastatic breast cancer [8]. In the DAISY trial, a phase 2 trial in patients with metastatic breast cancer with variable HER2 expression, tumor responses to T-DXd were observed in 27 of 72 patients (37.5%) with HER2-low and 11 of 37 patients (29.7%) classified as IHC 0 [29]. These initial promising results for the efficacy of T-DXd for the treatment of HER2-low advanced breast cancer were confirmed in the DESTINY-Breast04 study; patients with HER2-low breast cancer treated with T-DXd had a median progression-free survival of 10.1 months, compared with 5.4 months for patients treated with physician’s choice of chemotherapy (hazard ratio 0.51, p < 0.001) [12]. This led to T-DXd being the first targeted therapy approved for the treatment of HER2-low metastatic/unresectable breast cancer [13, 14].

Furthermore, the DESTINY-Breast06 study has confirmed efficacy of T-DXd in patients with HER2-low or HER2-ultralow metastatic breast cancer following ≥ 1 endocrine-based therapy. Patients in the intent-to-treat population (consisting of patients with hormone receptor–positive HER2-low and HER2-ultralow disease) treated with T-DXd had a median progression-free survival of 13.2 months, compared with 8.1 months for patients treated with physician’s choice of chemotherapy (hazard ratio 0.64, p < 0.001) [17]. The efficacy of new HER2-directed therapies in patients with HER2-ultralow breast cancer highlights the importance of accurate scoring at the HER2-ultralow cutoff (IHC 0 absent membrane staining vs. IHC 0 with membrane staining, 1+, 2+, 3+), as the results from DESTINY-Breast06 imply that any level of HER2 expression above IHC 0 with membrane staining is clinically relevant. Whether T-DXd could be effective in breast cancer patients with IHC 0 absent membrane staining tumors is currently unknown and is being investigated as part of the phase 3b DESTINY-Breast15 clinical trial, which is designed to evaluate T-DXd in patients with hormone receptor–positive or hormone receptor–negative HER2-low or HER2 IHC 0 metastatic breast cancer [30].

Historical scoring by real-world pathologists at the HER2-low cutoff (IHC 0 vs. IHC 1+, 2+, 3+) showed substantial agreement (κ = 0.72) with central consensus scores with OPA of 90%. This supports the use of historical HER2 IHC scores for informing reliable treatment decisions. Of note, these real-world testing data were captured before HER2-low was established as a clinically actionable cutoff, when there was less emphasis on distinguishing between IHC 0 and IHC 1+. Additionally, since HER2-low/ultralow scoring is relevant only in metastatic disease, clinical laboratories may assess only a limited number of samples over a given time period. The lower agreement across all HER2 scores (κ = 0.60) compared with agreement among central pathologists (κ = 0.69), and the presence of inter-laboratory variability, suggest that there is room for improvement in scoring reproducibility (e.g., with experience, increased awareness of the clinical importance of HER2-low and/or training) [25]. It should be noted that, by using central consensus as the reference score, a local result could be classified as discordant even if it agreed with one of the central pathologists (in cases where only 2/3 central pathologists agreed).

A previous study of scoring concordance of HER2 IHC samples found low accuracy at the HER2-low cutoff, reporting that pathologists in the study had not been told that the HER2-low cutoff would be assessed; many of these pathologists said that in retrospect, they would have examined samples with low HER2 expression more closely if they had been informed [22]. This suggests that increasing awareness of the clinical importance of HER2-low scoring may lead to improved focus and accuracy for scoring this cutoff. Emerging guidance around scoring the HER2-low cutoff is now available [5, 6], and new computational pathology/artificial intelligence tools are demonstrating utility in assisting pathologists to control for pre-analytical artefacts and score HER2, including at the HER2-low cutoff, all of which may improve scoring reproducibility [31, 32]. Altogether, increased awareness, updated guidelines, and the development of assistive tools are likely to improve reproducibility of scoring at the HER2-low cutoff beyond that seen in the historical, locally scored data in this study.

In the abovementioned HER2 IHC concordance study, variability in the HER2 staining assay was introduced because each participating laboratory used its own standard method for staining HER2 [22]; this may have contributed to lack of concordance between raters. In the current study, only samples stained with the VENTANA HER2 (4B5) assay according to the manufacturer’s instructions were used. Following the approval of T-DXd for treatment of HER2-low disease, the VENTANA HER2 (4B5) assay is now approved as the companion diagnostic test for T-DXd for HER2-low scoring in breast cancer, in addition to the previous approval of the assay for distinguishing HER2-positive from HER2-negative breast cancer [7, 18]. Standardization of staining assays between laboratories is likely to aid further the reproducible and reliable scoring of the HER2-low cutoff.

Accurate identification of patients with HER2-low tumors will benefit an appreciable proportion of patients with metastatic breast cancer because approximately 60% of primary/metastatic breast cancers traditionally categorized as HER2-negative are HER2-low, according to an international study of 3689 samples [33]. A retrospective study on the prevalence and outcomes of HER2-low breast cancer reported moderate concordance when historical HER2 samples were rescored by pathologists trained in HER2-low scoring. This study found that approximately two-thirds of patients with historically HER2-negative unresectable/metastatic breast cancer may stand to benefit from HER2-low directed treatments, underlining the importance of training pathologists in scoring HER2-low [26].

Our analysis revealed substantial agreement between pathologists for tumor samples obtained by core-needle biopsy or other biopsies (κ = 0.75, n = 415). Given that the majority of samples in the metastatic breast cancer setting are obtained from biopsies [18], our data support reproducible scoring of HER2-low in samples relevant to clinical practice. Moderate agreement was observed for tumor samples obtained by excisions or resection (κ = 0.52, n = 69). Only a relatively small number of samples obtained by excision or resection were evaluated, so we are not able to draw strong conclusions about the suitability of this sample collection method. Of note, intratumor heterogeneity may lead to large variability in HER2 expression across a sample, an issue that is more apparent in HER2-low disease [34], which may explain the lower scoring agreement in tissue samples obtained by excision or resection.

A limitation of our study is that historical agreement at the HER2-ultralow cutoff (IHC 0 absent membrane staining vs. IHC 0 with membrane staining, 1+, 2+, 3+) could not be assessed. Differentiating IHC 0 absent membrane staining from IHC 0 with membrane staining was not part of standard clinical practice at the time of historical sample scoring, so data were not in real-world pathology reports. Our data indicate substantial agreement between three central pathologists scoring at the HER2-ultralow cutoff (κ = 0.68), indicating that scoring this cutoff is reproducible. However, a larger group of pathologists would allow for a more robust assessment of scoring agreement and further validate the reproducibility of the scoring system. Further investigation of scoring accuracy at this cutoff by pathologists in the clinical setting will be important if patients with HER2-ultralow breast cancer gain new treatment options as a result of the DESTINY-Breast06 clinical trial.

Conclusions

Our data demonstrate that it is feasible to reproducibly score the HER2-low and HER2-ultralow cutoff, provided that recommendations made in guidance for HER2-low scoring are implemented and that pathologists are trained in scoring HER2-low and HER2-ultralow. Real-world scoring of HER2-low showed substantial agreement with central testing, supporting the use of historical HER2 IHC scores to inform reliable treatment decisions.

Data availability

No datasets were generated or analysed during the current study.

Abbreviations

ASCO:

American Society of Clinical Oncology

CAP:

College of American Pathologists

CE:

Conformité Européene

CI:

Confidence interval

HER2:

Human epidermal growth factor receptor 2

IHC:

Immunohistochemistry

ISH:

In situ hybridization

IVDD:

In Vitro Diagnostic Medical Device Directive

NPA:

Negative percent agreement

OPA:

Overall percent agreement

PPA:

Positive percent agreement

T-DXd:

Trastuzumab deruxtecan

VENTANA HER2 (4B5):

VENTANA anti-HER2/neu (4B5) Rabbit Monoclonal Primary Antibody

References

  1. Wolff AC, Hammond ME, Schwartz JN, Hagerty KL, Allred DC, Cote RJ, et al. American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. Arch Pathol Lab Med. 2007;131:18–43.

  2. Wolff AC, Hammond ME, Allison KH, Harvey BE, Mangu PB, Bartlett JM, et al. Human epidermal growth factor receptor 2 testing in breast cancer: American Society of Clinical Oncology/College of American Pathologists clinical practice guideline focused update. J Clin Oncol. 2018;36:2105–22.

  3. Slamon DJ, Leyland-Jones B, Shak S, Fuchs H, Paton V, Bajamonde A, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med. 2001;344:783–92.

  4. Gennari A, André F, Barrios CH, Cortés J, de Azambuja E, DeMichele A, et al. ESMO Clinical Practice Guideline for the diagnosis, staging and treatment of patients with metastatic breast cancer. Ann Oncol. 2021;32:1475–95.

  5. Wolff AC, Somerfield MR, Dowsett M, Hammond ME, Hayes DF, McShane LM, et al. Human epidermal growth factor receptor 2 testing in breast cancer: ASCO–College of American Pathologists guideline update. J Clin Oncol. 2023;41:3867–72.

  6. Tarantino P, Viale G, Press MF, Hu X, Penault-Llorca F, Bardia A, et al. ESMO expert consensus statements (ECS) on the definition, diagnosis, and management of HER2-low breast cancer. Ann Oncol. 2023;34:645–59.

  7. PATHWAY anti-HER2/neu (4B5) rabbit monoclonal primary antibody package insert [Internet]. 2022 [cited 2024 Feb 23]. Available from: https://elabdoc-prod.roche.com/eLD/api/downloads/fc569ff5-2236-ea11-fc90-005056a71a5d?countryIsoCode=us

  8. Modi S, Park H, Murthy RK, Iwata H, Tamura K, Tsurutani J, et al. Antitumor activity and safety of trastuzumab deruxtecan in patients with HER2-low-expressing advanced breast cancer: results from a phase Ib study. J Clin Oncol. 2020;38:1887–96.

  9. Mosele F, Deluche E, Lusque A, Le Bescond L, Filleron T, Pradat Y, et al. Trastuzumab deruxtecan in metastatic breast cancer with variable HER2 expression: the phase 2 DAISY trial. Nat Med. 2023;29:2110–20.

  10. Nakada T, Sugihara K, Jikoh T, Abe Y, Agatsuma T. The latest research and development into the antibody-drug conjugate, [fam-] trastuzumab deruxtecan (DS-8201a), for HER2 cancer therapy. Chem Pharm Bull (Tokyo). 2019;67:173–85.

  11. Ogitani Y, Aida T, Hagihara K, Yamaguchi J, Ishii C, Harada N, et al. DS-8201a, a novel HER2-targeting ADC with a novel DNA topoisomerase I inhibitor, demonstrates a promising antitumor efficacy with differentiation from T-DM1. Clin Cancer Res. 2016;22:5097–108.

  12. Modi S, Jacot W, Yamashita T, Sohn J, Vidal M, Tokunaga E, et al. Trastuzumab deruxtecan in previously treated HER2-low advanced breast cancer. N Engl J Med. 2022;387:9–20.

  13. European Medicines Agency (EMA). ENHERTU: summary of product characteristics [Internet]. 2024 [cited 2024 Sep 25]. Available from: https://www.ema.europa.eu/en/documents/product-information/enhertu-epar-product-information_en.pdf

  14. U.S. Food and Drug Administration (FDA). ENHERTU (fam-trastuzumab deruxtecan-nxki): highlights of prescribing information [Internet]. 2024 [cited 2024 Sep 25]. Available from: www.accessdata.fda.gov/drugsatfda_docs/label/2024/761139s028lbl.pdf

  15. Daiichi-Sankyo. Press release. ENHERTU® approved in Japan as first HER2 directed therapy for patients with HER2 low metastatic breast cancer [Internet]. 2023 [cited 2024 Apr 5]. Available from: https://www.daiichisankyo.com/files/news/pressrelease/pdf/202303/20230327_E.pdf

  16. AstraZeneca. Press release. Enhertu approved in China as the first HER2-directed therapy for patients with HER2-low metastatic breast cancer [Internet]. 2023 [cited 2024 Apr 5]. Available from: https://www.astrazeneca.com/media-centre/press-releases/2023/enhertu-approved-in-china-as-the-first-her2-directed-therapy-for-patients-with-her2-low-metastatic-breast-cancer.html

  17. Bardia A, Hu X, Dent R, Yonemori K, Barrios CH, O’Shaughnessy JA, et al. Trastuzumab deruxtecan after endocrine therapy in metastatic breast cancer. N Engl J Med. 2024;391:2110–22.

  18. Garrido C, Manoogian M, Ghambire D, Lucas S, Karnoub M, Olson MT, et al. Analytical and clinical validation of PATHWAY anti-HER-2/neu (4B5) antibody to assess HER2-low status for trastuzumab deruxtecan treatment in breast cancer. Virchows Arch. 2024;484:1005–14.

  19. Australian approval for use of VENTANA HER2 (4B5) assay for HER2 IHC [Internet]. 2023 [cited 2024 May 9]. Available from: https://www.ebs.tga.gov.au/servlet/xmlmillr6?dbid=ebs/PublicHTML/pdfStore.nsf%26docid=421354%26agid=%28PrintDetailsPublic%29%26actionid=1

  20. Roche. Press release. Roche obtains CE Mark for first companion diagnostic to identify patients with HER2-low metastatic breast cancer eligible for ENHERTU [Internet]. Diagnostics. 2024 [cited 2024 May 9]. Available from: https://diagnostics.roche.com/global/en/news-listing/2024/roche-obtains-ce-mark-for-first-companion-diagnostic-to-identify-patients-with-her2-low-metastatic-breast-cancer-eligible-for-enhertu.html

  21. Japanese Pharmaceuticals and Medical Devices Agency. List of in vitro companion diagnostics or medical devices (CDx products) approved in Japan [Internet]. 2024 [cited 2024 May 13]. Available from: https://www.pmda.go.jp/files/000266999.pdf

  22. Fernandez AI, Liu M, Bellizzi A, Brock J, Fadare O, Hanley K, et al. Examination of low ERBB2 protein expression in breast cancer tissue. JAMA Oncol. 2022;8:607–10.

  23. Zaakouk M, Quinn C, Provenzano E, Boyd C, Callagy G, Elsheikh S, et al. Concordance of HER2-low scoring in breast carcinoma among expert pathologists in the united Kingdom and the Republic of Ireland – on behalf of the UK National coordinating committee for breast pathology. Breast. 2023;70:82–91.

  24. Karakas C, Tyburski H, Turner BM, Wang X, Schiffhauer LM, Katerji H, et al. Interobserver and interantibody reproducibility of HER2 immunohistochemical scoring in an enriched HER2-low–expressing breast cancer cohort. Am J Clin Pathol. 2023;159:484–91.

  25. Rüschoff J, Penner A, Ellis IO, Hammond MEH, Lebeau A, Osamura RY, et al. HER2-13 proficiency assessment of HER2-low breast cancer scoring with the Ventana PATHWAY 4B5 and Dako HercepTest HER2 assays and the impact of pathologist training. Cancer Res. 2023;83(Suppl 5):HER2–13. (Abstract).

  26. Viale G, Basik M, Niikura N, Tokunaga E, Brucker S, Penault-Llorca F, et al. Retrospective study to estimate the prevalence and describe the clinicopathological characteristics, treatments received, and outcomes of HER2-low breast cancer. ESMO Open. 2023;8:101615.

  27. Scott M, Vandenberghe ME, Scorer P, Boothman A-M, Barker C. Prevalence of HER2 low in breast cancer subtypes using the VENTANA anti-HER2/neu (4B5) assay. J Clin Oncol. 2021;39(Suppl. 15):1021 (Abstract).

  28. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33:159–74.

  29. Dieras VC, André F, Lusque A, Pistilli B, Bachelot T, Pierga J, et al. Trastuzumab deruxtecan (T-DXd) for advanced breast cancer (ABC) patients (pts), regardless of HER2 status: a phase II study with biomarkers analysis (DAISY) – update analysis with longer follow-up. ESMO Open. 2024;9(Suppl 4):4. (Abstract).

  30. Modi S, Salgado R, Guarneri V, Alagappan D, Dennis N, Newell AH et al. An open-label, interventional, multicenter study of trastuzumab deruxtecan monotherapy in patients with unresectable and/or metastatic HER2-low or HER2 immunohistochemistry 0 breast cancer: DESTINY-Breast15. Cancer Res. 2024;84(Suppl. 9):PO2-19-06 (Abstract).

  31. Krishnamurthy S, Schnitt S, Vincent-Salomon A, Canas-Marques R, Colon E, Kantekure K, et al. Fully automated artificial intelligence solution for human epidermal growth factor receptor 2 immunohistochemistry scoring in breast cancer: a multireader study. JCO Precis Oncol. 2024;8:e2400353.

  32. Brevet M, Li Z, Parwani A. Computational pathology in the identification of HER2-low breast cancer: opportunities and challenges. J Pathol Inform. 2023;15:100343.

  33. Schettini F, Chic N, Brasó-Maristany F, Paré L, Pascual T, Conte B, et al. Clinical, pathological, and PAM50 gene expression features of HER2-low breast cancer. NPJ Breast Cancer. 2021;7:1.

  34. Hou Y, Nitta H, Li Z. HER2 intratumoral heterogeneity in breast cancer, an evolving concept. Cancers (Basel). 2023;15:2664.

Download references

Acknowledgements

Medical writing support, under the direction of the authors, was provided by Georgina Stooke-Vaughan, PhD, of Helios Medical Communications, part of Helios Global Group, and was funded by AstraZeneca in accordance with Good Publication Practice (GPP) guidelines (http://www.ismpp.org/gpp-2022).The authors would like to thank Marianne Ratcliffe for providing comments on the manuscript and CellCarta for provision of training and central pathologist scoring results.

Funding

This study is sponsored by AstraZeneca. In March 2019, AstraZeneca entered into a global development and commercialization collaboration agreement with Daiichi Sankyo for trastuzumab deruxtecan (T-DXd; DS-8201).

Author information

Authors and Affiliations

  1. Oncology R&D, AstraZeneca, Postępu 14, Warsaw, 02-676, Poland

    Agata Wróbel

  2. R&D, AstraZeneca, Cambridge, UK

    Michel Vandenberghe, Marietta Scott, Frances Jones, Tsuyoshi Matsuo, Anne-Marie Boothman, Jessica Whiteley & Craig Barker

Authors
  1. Agata Wróbel
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Michel Vandenberghe
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Marietta Scott
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Frances Jones
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Tsuyoshi Matsuo
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Anne-Marie Boothman
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Jessica Whiteley
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Craig Barker
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

All authors participated in the participated in conception and design, acquisition of data, analysis and interpretation of data, critical revision of the manuscript for important intellectual content, and statistical analysis.

Corresponding author

Correspondence to Agata Wróbel.

Ethics declarations

Ethics approval and consent to participate

AstraZeneca has a governance framework and processes to ensure that commercial sources have appropriate patient consent and ethical approval in place for collection of samples for research purposes, including use by for-profit companies. The AstraZeneca Biobank in the UK is licensed by the Human Tissue Authority (Licence No. 12109) and has National Research Ethics Service Committee (NREC) approval as a Research Tissue Bank (RTB) (REC No 22/NW/0102), which covers the use of the samples for this project.

Consent for publication

Not applicable.

Competing interests

All authors are employees of AstraZeneca. MV, MS, FJ, TM, A-MB, JW, and CB hold shares in AstraZeneca.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wróbel, A., Vandenberghe, M., Scott, M. et al. Accuracy of human epidermal growth factor receptor 2 (HER2) immunohistochemistry scoring by pathologists in breast cancer, including the HER2-low cutoff. Diagn Pathol 20, 35 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13000-025-01624-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13000-025-01624-3

Keywords

Diagnostic Pathology

ISSN: 1746-1596

Contact us