Skip to main content
Download PDF

Prognostic impact of HER2 biomarker levels in trastuzumab-treated early HER2-positive breast cancer

Breast Cancer Research volume 26, Article number: 24 (2024) Cite this article

Abstract

Background

Overexpression of human epidermal growth factor receptor 2 (HER2) caused by HER2 gene amplification is a driver in breast cancer tumorigenesis. We aimed to investigate the prognostic significance of manual scoring and digital image analysis (DIA) algorithm assessment of HER2 copy numbers and HER2/CEP17 ratios, along with ERBB2 mRNA levels among early-stage HER2-positive breast cancer patients treated with trastuzumab.

Methods

This retrospective study comprised 371 early HER2-positive breast cancer patients treated with adjuvant trastuzumab, with HER2 re-testing performed on whole tumor sections. Digitized tumor tissue slides were manually scored and assessed with uPath HER2 Dual ISH image analysis, breast algorithm. Targeted ERBB2 mRNA levels were assessed by the Xpert® Breast Cancer STRAT4 Assay. HER2 copy number and HER2/CEP17 ratio from in situ hybridization assessment, along with ERBB2 mRNA levels, were explored in relation to recurrence-free survival (RFS).

Results

The analysis showed that patients with tumors with the highest and lowest manually counted HER2 copy number levels had worse RFS than those with intermediate levels (HR = 2.7, CI 1.4–5.3, p = 0.003 and HR = 2.1, CI 1.1–3.9, p = 0.03, respectively). A similar trend was observed for HER2/CEP17 ratio, and the DIA algorithm confirmed the results. Moreover, patients with tumors with the highest and the lowest values of ERBB2 mRNA had a significantly worse prognosis (HR = 2.7, CI 1.4–5.1, p = 0.003 and HR = 2.8, CI 1.4–5.5, p = 0.004, respectively) compared to those with intermediate levels.

Conclusions

Our findings suggest that the association between any of the three HER2 biomarkers and RFS was nonlinear. Patients with tumors with the highest levels of HER2 gene amplification or ERBB2 mRNA were associated with a worse prognosis than those with intermediate levels, which is of importance to investigate in future clinical trials studying HER2-targeted therapy.

Background

Human epidermal growth factor receptor 2 (HER2) is encoded by the oncogene ERBB2 and is overexpressed in around 15% of all primary breast cancers [1, 2]. In clinical routine, HER2 status is determined with HER2 protein expression analyzed by immunohistochemistry (IHC) and HER2 gene amplification assessed by a DNA probe integrated into an in situ hybridization (ISH) detection system. HER2 was discovered decades ago, but the same diagnostic methods are still in use despite being time-consuming and hampering reproducibility problems [2,3,4]. Several other techniques for determining HER2 status at either protein, RNA or DNA levels are available but have not yet reached routine practice [5]. In addition, HER2-positive breast cancer patients are treated similarly despite individual variations in HER2 copy number and HER2/Chromosome enumeration probe 17 (CEP17) ratio levels.

Unlike HER2 diagnostic methods, treatment options for HER2-positive breast cancer patients have evolved rapidly over the last decades [6,7,8,9,10,11]. Today, high-risk early-stage HER2-positive breast cancer patients are most often recommended neoadjuvant treatment with a combination of monoclonal antibodies that bind to the extracellular domain of the HER2 receptor (e.g., trastuzumab and pertuzumab) together with chemotherapy [3, 11,12,13]. Post-surgery, patients usually receive additional trastuzumab, but in case of residual disease, the antibody–drug conjugate (ADC) trastuzumab emtansine (T-DM1), which combines trastuzumab with a cytotoxic drug called emtansine or DM1 is recommended [3, 7, 14]. Recently, new, more effective ADCs for HER2-positive breast cancer, such as trastuzumab deruxtecan (T-DXd), have been introduced [15, 16]. In addition, T-DXd demonstrated activity against HER2 low tumors (i.e., HER2 IHC 1+ and non-amplified HER2 IHC 2+) [17]. In the era of new-generation ADCs, it is essential to improve the definition of HER2 testing. Therefore, there is an interest in investigating HER2 biomarker levels in relation to outcomes among HER2-positive breast cancer. In addition, local and regional variations in HER2 positivity rates are evident, and it is essential to limit false-negative and false-positive HER2 results through standardization and quality control [18,19,20].

One solution to standardize HER2 assessment and reduce HER2 analysis time is through digital image analysis (DIA). Current studies mainly focus on HER2 IHC scoring [21,22,23]. Apart from a few artificial intelligence (AI)-based models that have been developed to perform IHC scoring from IHC-stained slides, another model has demonstrated the capability to predict HER2 status directly from hematoxylin and eosin (H&E)-stained slides [24, 25]. Recently, an ISH digital image analysis algorithm that could aid pathologists in investigating HER2 at a DNA level was developed [26,27,28]. Since ISH scoring is the most time-consuming procedure in HER2 diagnostics and is hampered by variability between readers, decision aids are essential for pathologists and patients.

This study aimed to systematically re-assess HER2 biomarkers including HER2 ISH and ERBB2 mRNA, and correlate them to outcomes in trastuzumab-treated early HER2-positive breast cancer patients. We aimed to investigate continuous levels of the biomarkers and their prognostic potential to discriminate between patients with therapy-resistant disease and those with good outcomes.

Materials and methods

Stockholm HER2 cohort study design

The Stockholm HER2 cohort is a retrospective identified cohort comprising patients diagnosed with HER2-positive early-stage primary breast cancer at the Department of Clinical Pathology and Cancer Diagnostics at the Karolinska University Hospital, Stockholm, Sweden, between 2006 and 2014. The pathology laboratory information system was searched to identify all HER2-analyzed tumors at the Karolinska University Hospital and to extract clinicopathological data (tumor size, axillary lymph node status, estrogen receptor (ER), progesterone receptor (PR), Ki67, HER2 by IHC and ISH). In addition, treatment information and at least 5-year follow-up data were extracted from medical records. The outcome variable was recurrence-free survival (RFS), defined as the time from the pathology-verified diagnosis until the time of recurrence or death by any cause, according to the STEEP criteria [29]. Exclusion criteria were as follows: HER2-negative breast cancer, previous ipsilateral breast cancer diagnosis, bilateral breast cancer, stage IV disease at diagnosis, recurrence before HER2-targeted therapy, no HER2-targeted treatment, lack of follow-up data, no HER2 status on untreated tumor tissue, no invasive tumor or duplicates (Fig. 1).

Fig. 1
figure 1

CONSORT diagram of the Stockholm HER2 cohort. ASCO American Society of Clinical Oncology, BC breast cancer, CAP College of American Pathologists, CNB core needle biopsy, DCIS ductal cancer in situ, ISH in situ hybridization, RFS recurrence-free survival

Full size image

After exclusions, the study cohort comprised 474 primary tumors with available original HER2 status analyzed on either core needle biopsy or surgical resection specimen prior to treatment. Archived tumor tissue material from the Stockholm medical biobank was retrieved, sectioned and stained with H&E; stained and analyzed for HER2 IHC and HER2 ISH; and analyzed for ERBB2 mRNA using Xpert Breast Cancer STRAT4 (CE-IVD. In vitro diagnostic medical device. Not available in all countries. Not available in the U.S) assay. After further exclusion of patients with HER2 negative tumors on re-assessment with the ASCO guidelines updated 2018 [4] and those with incomplete data on one of HER2 gene amplification, ERBB2 mRNA or outcome data (RFS), a total of 434 patients were available for statistical analysis. Patients had been treated with trastuzumab either in the neoadjuvant setting (14.5%) or the adjuvant setting (85.5%). In this study, we focused on the adjuvant-treated part of the cohort, thus including 371 primary tumors for analysis (Fig. 1). This study was performed and reported to the greatest extent in accordance with the REMARK guidelines [30]. The biospecimen reporting for improved study quality (BRISQ) criteria for this cohort are shown in the Additional file 1: Table S1 [31].

HER2 immunohistochemistry and in situ hybridization

In the original biomarker assessment, fluorescent or chromogen ISH was routinely performed for all HER2 IHC score 2+  and 3+ tumors. As the fluorescent ISH signals fade over time, all tumors with available tumor tissue were re-tested for HER2 (from December 2019 to June 2020) to accurately compare HER2 status across the cohort. New parallel 4 μm tumor tissue whole sections from the archived formalin-fixed paraffin-embedded (FFPE) material were stained with the PATHWAY anti-HER2/neu (4B5) Rabbit Monoclonal Primary Antibody (Roche Diagnostics International, Rotkreutz, Switzerland) as described by the manufacturer (BenchMark ULTRA IHC/ISH Staining Module, Ventana Medical Systems, Inc., Arizona, USA). Similarly, ISH for HER2 was performed on a parallel 4 μm tumor tissue section. HER2 dual-probe ISH staining utilized VENTANA HER2 Dual ISH DNA Probe Cocktail assay (Roche Diagnostics International, Rotkreutz, Switzerland) together with VENTANA Silver ISH DNP Detection kit and VENTANA Red ISH DIG detection kit as described by the manufacturer (BenchMark ULTRA IHC/ISH Staining Module, Ventana Medical Systems, Inc., Arizona, USA). For accurate histopathological assessment of HER2 status, all tumors also included a parallel whole tumor section stained with H&E.

Manual HER2 scoring

All new tumor slides (including H&E, HER2 IHC and HER2 ISH) were scanned at 40X using a NanoZoomer XR (Hamamatsu Photonics K.K., Japan) digital slide scanner. The NDP.view2 (Hamamatsu Photonics K.K., Japan) viewing software was used to view the whole slide images. All re-tested HER2 sections were evaluated by at least two pathologists (CR resident pathologist and JH or SR, both board-certified breast pathologists). The manual scoring of IHC and ISH was performed at 40-60X on digitized whole slide images or by brightfield light microscopy for a few cases. According to the Swedish national guidelines, the number of HER2 signals and CEP17 signals per cell was counted in 20 tumor cells in two separate areas of the invasive tumor. The average HER2 signals/cell (hereafter copy number) and HER2/CEP17 ratio were reported for each case. HER2 IHC score 0 to 3+ was defined as follows: 0 as no or incomplete faint/weak membrane staining in ≤ 10%; 1+ as incomplete faint/weak membrane staining in > 10%; 2+ as weak to moderate complete membrane staining in > 10%, or complete intense staining in ≤ 10%; 3+ as complete intense membrane staining in > 10% of tumor cells [4, 32, 33]. IHC 0–1+ was defined as negative, IHC 2+ as equivocal requiring reflex test with ISH, and IHC 3+ as positive [33].

The Swedish national guidelines of 2020 were used for HER2 re-testing and are in line with the ASCO/CAP 2018 guidelines [4, 32, 33]; tumors assessed as IHC 3+ or 2+ together with a HER2/CEP17 ratio ≥ 2 with HER2 copy number ≥ 4 signals/cell (ASCO/CAP dual ISH group 1) or HER2/CEP17 ratio < 2 with HER2 copy number ≥ 6 signals/cell (ISH group 3) were considered as HER2-positive. Thus, ISH group 5 was considered negative (HER2/CEP17 ratio < 2.0 and HER2 copy number < 4.0 signals/cell). The ASCO/CAP dual ISH groups 2 and 4 are the less common clinical scenarios and were assessed according to the ASCO CAP algorithm [32].

HER2 scoring by digital image analysis

All tumor slides were scanned at 40X for ISH using the VENTANA DP 200 slide scanner (Roche Diagnostics International, Rotkreuz, Switzerland). The images were analyzed with the uPath HER2 Dual ISH image analysis, breast algorithm (Roche Diagnostics International, Rotkreuz, Switzerland) between June 2022 to December 2022 using Roche uPath enterprise software (Roche Diagnostics International, Rotkreuz, Switzerland). The DIA algorithm scoring was performed (by CR) with a substantial wash-out period of at least 18 months from the manual scoring. uPath HER2 Dual ISH image analysis, breast algorithm is a partly automatic DIA algorithm that imitates a pathologist’s ISH assessment and has been described previously [26]. Briefly, the pathologist identified two regions of interest (ROIs) using the provided heatmaps within the viewer and excluded non-invasive areas and areas with low HER2 amplification. Thereafter, the DIA algorithm selected 20 cells per ROI and automatically output each cell's HER2 copy number count and a CEP17 count. Finally, the pathologist confirmed each cell count and if needed, deleted unacceptable cells and chose new countable tumor cells. Eventually, the DIA algorithm summarized all counted cells and presented an average HER2 copy number and HER2/CEP17 ratio per analyzed case. The HER2 scoring by the DIA algorithm was performed and reported separately from the manual HER2 scoring results, ERBB2 mRNA results and the outcome endpoint.

ERBB2 mRNA by RT-PCR

To measure ERBB2 mRNA, the CE-IVD marked (In vitro diagnostic medical device. Not available in all countries. Not available in the U.S.) analysis Xpert® Breast Cancer STRAT4 Assay (Cepheid, Sunnyvale, CA, USA) based on real-time polymerase chain reaction (RT-PCR) was performed between December 2019 and June 2020. From archived FFPE material, macrodissection (with > 90% tumor content) of each invasive tumor was performed and a 10 μm tumor tissue section cut and placed into tubes. The tumor material was prepared according to the manufacturer’s instructions (Xpert Breast Cancer STRAT4® Assay, Cepheid, Sunnyvale, CA, USA) [34]. Briefly, the samples were treated with FFPE lysis reagent and Proteinase K and incubated at 80 °C for 30 min. Next, the content was diluted with ≥ 95% ethanol, and the lysate was added to the closed-system STRAT4® cartridge and analyzed at the semi-automatic GeneXpert (GX system) which performed isolation of RNA, amplification and real-time detection of mRNA. Cycle thresholds (Ct) were determined for ERBB2 and the endogenous reference gene called Cytoplasmic FMR1-Interacting Protein 1 (CYFIP1). The ERBB2 mRNA results were presented as a delta Ct (dCt) value, which was defined as the CYFIP1 Ct minus the ERBB2 Ct. ≥ − 1.0 dCt was the predetermined cutoff referred to as an ERBB2 mRNA positive result. The cutoff was provided by the manufacturer and was based on previous studies [34,35,36]. The ERBB2 mRNA analysis was performed blinded from the manual HER2 assessment and the DIA algorithm scoring.

Statistical analysis

Methods to search for optimal cutoffs were applied for the study cohort n = 371 according to Fig. 1, and patients were classified as low, intermediate or high for each investigated biomarker. Specifically, cutoff determination for HER2 copy number (manual and DIA algorithm), HER2/CEP17 ratio (manual and DIA algorithm) and ERBB2 mRNA was investigated by positional scanning (PS) analysis using Cutoff Finder [37]. Cutoff Finder selects the optimal cutoff associated with the lowest p-value for RFS. In addition, subpopulation treatment effect pattern plot (STEPP) analysis suggested a second cutoff, which was also defined by positional scanning as described previously [38,39,40,41]. For the STEPP analysis, which examines the relationship between a continuous covariate and the probability of survival at a predefined time point, a sliding-window approach was used to define several overlapping subpopulations of patients [41]. The optimal values for generating the patient subpopulations were defined by the utility function balance_patients in R package stepp. (This divided the cohort into 9–11 subgroups, with approximately 33–41 patients in each group, depending on the biomarker in question.) The reverse Kaplan–Meier (KM) estimate of the median RFS follow-up was used as the predefined time point in the STEPP analysis [42].

The low, intermediate and high groups of patients were used for survival analysis, which was performed with the R package survival [43] using RFS as the clinical endpoint. Specifically, KM estimates and univariate and multivariable Cox proportional hazards (PH) regression models were applied, and hazard ratios (HRs) with 95% confidence intervals (CIs) were estimated. The PH assumption was checked with the scaled Schoenfeld residuals. Continuous co-variables in the multivariable Cox regression models were tested for nonlinearity by plotting the Martingale residuals. Estrogen receptor status was found to violate the PH assumption, and thus, it was used as a stratification factor in the multivariable Cox regression models, which also included tumor size (categorical, ≤ 20 mm; > 20 to ≤ 50 mm; > 50 mm) and lymph node status (categorical, LN−; LN+). All statistical tests were two-sided. Significance was considered at a p < 0.05 level. All bioinformatics and statistical analyses including descriptive tumor characteristics were performed within the R computing environment version 4.2.2 (2022-10-31) or later, © 2022 by Posit Software, PBC.

Results

Patient characteristics

Results showed that 327 (88.1%) of the patients had tumors that were IHC 3+, 42 (11.3%) were IHC 2+ and 2 (0.5%) were IHC 1+ . The median average HER2 signals/cell by manual scoring was 9.8 signals/cell (range 2.3–21.6), and the median HER2/CEP17 ratio by manual scoring was 5.8 (range 1.4–18.8). Among all tumors, 226 (60.9%) were ER-positive (cutoff ≥ 10%), 140 (37.7%) were PR-positive (cutoff ≥ 10%) and 322 (86.8%) had high Ki67 (cutoff ≥ 20%). All patients were treated with adjuvant trastuzumab and 360 (97.0%) were treated with chemotherapy. During follow-up, 38 (10.2%) of the patients experienced a recurrence of which a majority 30 (78.9%) had a distant metastasis as the first recurrence. The median RFS time was 8.7 years (defined by the reverse Kaplan–Meier estimate) [42]. Complete tumor characteristics, treatment and outcome data of the study cohort (n = 371) are presented in the Additional file 1: Tables S2-5.

Prognostic levels of HER2 copy number and HER2/CEP17 ratio

To determine the prognostic significance of different levels of manual and DIA algorithm scoring of HER2 gene amplification, positional scanning with Cutoff Finder along with STEPP analysis were performed and presented in Figs. 2 and 3.

Fig. 2
figure 2

HER2 copy number investigated with subpopulation treatment effect pattern plot (STEPP) and survival analysis. STEPP investigated the relationship between manually- (A) and digital image analysis (DIA) algorithm-scored (C) mean HER2 copy number and the probability of being recurrence-free at the median recurrence-free survival (RFS) (8.7 years). The lower left corner in A, C shows the cutoff levels suggested by STEPP analysis and identified by positional scanning (PS). These cutoffs divided the cohort into three subgroups, namely low, intermediate and high. The locally estimated scatterplot smoothing curve fit is shown with the blue line. Kaplan–Meier estimates and multivariable Cox regression model of manually- (B) and DIA algorithm-scored (D) HER2 copy number related to RFS. The study cohort was divided into three subgroups according to the suggested optimal first and second cutoffs for the manually scored HER2 copy number and the DIA algorithm scored HER2 copy number as shown in A, C. *The multivariable Cox regression model was adjusted for lymph node status, tumor size and stratified by estrogen receptor status. Adjusted hazard ratios (HR) with 95% confidence interval (CI) are presented

Full size image
Fig. 3
figure 3

HER2/CEP17 ratio investigated with subpopulation treatment effect pattern plot (STEPP) and survival analysis. STEPP investigated the relationship between manually- (A) and digital image analysis (DIA) algorithm-scored (C) mean HER2/CEP17 ratio and the probability of being recurrence-free at the median recurrence-free survival (RFS) (8.7 years). The lower left corners in A, C show the cutoff levels suggested by STEPP analysis and identified by positional scanning (PS). These cutoffs divided the cohort into three subgroups, namely low, intermediate and high. The locally estimated scatterplot smoothing curve fit is shown with the blue line. Kaplan–Meier estimates and multivariable Cox regression model of manually- (B) and DIA algorithm-scored (D) HER2/CEP17 ratio related to RFS. The study cohort was divided into three subgroups according to the suggested optimal first and second cutoffs for the manually scored HER2/CEP17 ratio and the DIA algorithm scored HER2/CEP17 ratio as shown in A, C. *The multivariable Cox regression model for manual HER2/CEP17 ratio was adjusted for lymph node status, tumor size and stratified for estrogen receptor status. Adjusted hazard ratios (HR) with 95% confidence interval (CI) are presented

Full size image

For the manually scored HER2 copy number, cutoffs were suggested at 14.03 and 7.03 signals/cell in relation to RFS (log-rank p = 0.002), and the STEPP curve indicated that patients with tumors with intermediate levels of HER2 copy number were associated with superior outcomes (Figs. 2A, B and Additional file 1: Table S6). In multivariable Cox regression analysis, patients with tumors with high levels of HER2 copy number (≥ 14.03 signals/cell) showed a significantly worse RFS than patients with tumors with intermediate levels (HR = 2.7, CI 1.4–5.3, p = 0.003). In addition, the patients with tumors with the lowest levels of HER2 copy number (< 7.03 signals/cell) were associated with a significantly worse outcome than those with intermediate levels (HR = 2.1, CI 1.1–3.9, p = 0.03; Fig. 2B). The DIA algorithm estimation of HER2 copy number was higher than the manual scoring with a median DIA algorithm HER2 copy number of 22.3 signals/cell compared to the median manual HER2 copy number of 9.8 signals/cell. DIA algorithm showed just as manual counting, although with different cutoffs, that patients with tumors with intermediate HER2 copy numbers were associated with a significantly better prognosis than those with both high levels of HER2 copy number (HR = 1.9, CI 1.0–3.5, p = 0.04) and low levels (HR = 4.6, CI 1.6–13.1, p = 0.004; Fig. 2C, D).

For the manually scored HER2/CEP17 ratio, cutoffs were suggested at 11.32 and 3.01 in relation to RFS (log-rank p = 0.01), indicating that patients with tumors with both high and low ratio levels were associated with a worse prognosis than those with intermediate levels (Fig. 3A, B and Additional file 1: Table S6). Similarly, as the HER2 copy number levels showed, patients with tumors with the highest levels of HER2/CEP17 ratio (≥ 11.32) were associated with a worse prognosis when compared to intermediate levels (HR = 3.6, CI 1.4–9.2, p = 0.008), and for the patients with tumors with the lowest levels of HER2/CEP17 ratio (< 3.01) there was a trend for the worse outcome when compared to intermediate levels (HR = 2.1, CI 1.0–4.3, p = 0.05; Fig. 3B).

DIA algorithm scoring of HER2/CEP17 ratio confirmed the findings from the manual scoring but with different cutoffs. Patients with tumors with high HER2/CEP17 ratio levels showed a trend toward worse prognosis as compared to those with intermediate levels (HR = 1.8, CI 1.0–3.2, p = 0.06) and patients with tumors with low levels of HER2/CEP17 ratio had a significantly worse prognosis than those with intermediate levels (HR = 4.8, CI 1.9–12.1, p = 0.0009), (Fig. 3C, D). The HER2 DIA algorithm failed to scan or analyze six tumor slides. In addition, for each case with two ROIs (40 cells), a median of 3.0 (range 0–29) cells was changed by the resident pathologist (CR) because only parts of the cell nuclei were annotated as cells or non-tumor cells were counted.

Prognostic levels of ERBB2 mRNA

For STRAT4 ERBB2 mRNA positional scanning with Cutoff Finder along with STEPP analysis suggested cutoffs at 1.75 dCt and − 1.05 dCt in relation to RFS (log rank p = 0.001). Notably, the second cutoff was similar to the manufacturer’s set cutoff at − 1.0 dCt. The STEPP curve confirmed the previous pattern observed for HER2 copy number and HER2/CEP17 ratio, and patients with tumors with intermediate levels of ERBB2 mRNA were associated with significantly better survival outcomes (Fig. 4 and Additional file 1: Table S6). Patients with tumors with the highest ERBB2 mRNA levels (≥ 1.75 dCt) and lowest ERBB2 mRNA (< − 1.05 dCt) had a substantially worse RFS than those with tumors showing intermediate levels (HR = 2.7, CI 1.4–5.1, p = 0.003 and HR = 2.8, CI 1.4–5.5, p = 0.004, respectively).

Fig. 4
figure 4

ERBB2 mRNA investigated with subpopulation treatment effect pattern plot (STEPP) and survival analysis. STEPP investigated the relationship between mean ERBB2 mRNA (A) and the probability of being recurrence-free at the median recurrence-free survival (RFS) (8.7 years). The lower left corner shows the cutoff levels suggested by STEPP analysis and identified by positional scanning (PS). These cutoffs divided the cohort into three subgroups, namely low, intermediate and high. The locally estimated scatterplot smoothing curve fit is shown with the blue line. Kaplan–Meier estimates and multivariable Cox regression model of ERBB2 mRNA related to RFS (B). The study cohort was divided into three subgroups according to the suggested optimal first and second cutoffs for ERBB2 mRNA as shown in A. *The multivariable Cox regression model was adjusted for lymph node status, tumor size and stratified by estrogen receptor status. Adjusted hazard ratios (HR) with 95% confidence interval (CI) are presented

Full size image

Discussion

Recognizing the importance of precise levels of HER2 quantification is paramount with the emergence of novel treatments targeting HER2 in breast cancer. This understanding is crucial for including suitable candidates for clinical trials and, in the long run, for clinical utility. In this study, we systematically re-assessed HER2 copy number, HER2/CEP17 ratio and ERBB2 mRNA in a trastuzumab-treated breast cancer cohort and investigated the biomarkers in a continuous manner in relation to RFS. We showed that patients with tumors with intermediate levels of HER2 copy number, HER2/CEP17 ratio and ERBB2 mRNA had a superior prognosis compared to patients with tumors with both high and low levels of each biomarker. Furthermore, we confirmed the findings from manual scoring of HER2 gene amplification levels with a DIA algorithm scoring method.

Surprisingly, in this study none of the three HER2 biomarkers investigated by PS and STEPP analysis were linearly associated with outcome. Instead, we demonstrated that patients with tumors with the highest levels of HER2 copy number, HER2/CEP17 ratio and ERBB2 mRNA were consistently associated with worse RFS than those with intermediate levels of each biomarker. A few earlier studies have reported similar findings for individual HER2 biomarkers; STEPP analysis of HER2 copy number in the APHINITY trial showed that the lymph node-positive patients in the pertuzumab–trastuzumab treated group with tumors with the highest and the lowest levels of HER2 copy number had a lower treatment effect than those with intermediate levels of HER2 copy number [44]. Moreover, PS and STEPP analysis were used in two additional studies, one in the adjuvant and one in the metastatic setting. Both investigated HER2 at the protein level with the HERmark® assay (Monogram Biosciences Inc., South San Francisco, CA, USA) and found that patients treated with trastuzumab with very high tumor HER2 expression were associated with worse prognosis [38, 45]. In the present study, we investigated continuous assessments of HER2 in tumors at both the DNA and RNA levels, which to some extent, provides an internal validation of our findings. Moreover, to our knowledge, continuous assessments of ERBB2 mRNA by STEPP have not been published previously.

The recently published DAISY trial investigating the efficacy of T-DXd in metastatic breast cancer patients with variable HER2 expression has increased the interest in presenting detailed information on HER2 biomarker levels in clinical trials, and stakes have been raised to improve the speed of HER2 assessment in routine pathology [46].

Even though the DAISY trial enrolled patients based on HER2 IHC assessments and did not focus on detailed information on HER2 copy number and HER2 CEP/17 ratios, it may be interesting to investigate HER2 amplification levels in future studies since the ISH analysis is essential for clinical treatment decisions in HER2-positive breast cancer. However, ISH is a time-consuming procedure, and an automated digital assessment tool for ISH could provide valuable support. In the present study, the prognostic value of the DIA algorithm was similar to manual scoring.

As previously discussed by Griguolo et al. [47], cutoff levels for ADC treatment effect may differ depending on the type of molecule, indicating the importance of searching for relevant cutoff levels as in our study. For instance, Perez et al. showed that progression-free survival was superior in the T-DM1 group with higher ERBB2 mRNA (≥ median) compared to the trastuzumab group [48]. In addition, another trial of metastatic HER2-positive breast cancer confirmed a greater benefit for T-DM1 in contrast to physicians’ choice (typically trastuzumab-based regimes) in patients with tumors with high HER2 expression [49].

This hypothesis aligns with our results, indicating a worse prognosis in trastuzumab-treated HER2-positive patients with the highest HER2 biomarker levels and is also shown elsewhere [48, 50,51,52]. However, results have varied, mainly in the neoadjuvant setting, where high levels of HER2 gene amplification or HER2 gene expression have been associated with pathological complete response [53,54,55,56]. Subgroup analysis of adjuvant trastuzumab-treated HER2-positive patients showed a trend for less benefit in patients with the highest levels of HER2 copy number, but the results were insignificant [57].

In patients with residual disease after neoadjuvant HER2 target treatment, gene expression analysis of biomarkers revealed that high ERBB2 mRNA expression (> median) was associated with worse invasive disease-free survival within the group treated with trastuzumab in contrast to the T-DM1 group [50]. Considering our findings, we speculate that trastuzumab-treated patients with the highest biomarker levels may benefit from additional or more effective HER2 target treatment, e.g., ADCs.

Explanations as to why patients with tumors showing the highest and the lowest HER2 copy number, HER2/CEP17 ratio and ERBB2 mRNA had worse prognosis could possibly be found in the mechanistic investigations of HER2 therapy resistance, which are outside of the scope of this study and are therefore briefly described. Resistance to trastuzumab may occur in HER2-positive breast tumors because of HER2 molecular changes, including impaired or changed HER2 receptor epitope [58,59,60]. Trastuzumab may also be insufficient in blocking the heterodimerization of other HER family members, such as HER3, which is a resistance mechanism that could be overcome by adding the monoclonal antibody pertuzumab that binds to another part of the extracellular domain of HER2 [61, 62]. Other described causes of trastuzumab resistance are altered or alternative HER2 downstream signaling pathways, changed immune-related processes, changed metabolic processes, tumor cell plasticity, altered angiogenesis or intratumoral heterogeneity [63,64,65,66,67,68,69]. Interestingly, in vitro studies have shown that tumor cells with high expression of HER2 had an effective transport of the receptor back to the cell surface compared to cells with low HER2 expression, and one could speculate that this phenomenon might impact trastuzumab efficacy in tumors expressing the highest levels of HER2 [70].

Moreover, HER2-positive tumors with the lowest levels of HER2 gene amplification or HER2 expression may be insensitive to trastuzumab, and intratumor heterogeneity has been shown to impact trastuzumab efficacy [68, 71,72,73]. Lastly, another simpler explanation might be that patients with tumors with the highest levels of HER2 copy number, HER2/CEP17 ratio and ERBB2 mRNA could have been underdosed with trastuzumab.

Biological reasons as to why trastuzumab-treated patients with highly expressed or amplified HER2-positive tumors have a worse prognosis than those treated with T-DM1 are probably based on the linked cytotoxic agent of T-DM1 but are not yet thoroughly investigated [74]. It has been shown that T-DM1 keep the biological mechanisms of trastuzumab and that patients who did not have an effect of trastuzumab and taxanes might still respond to T-DM1 [14, 50, 75]. Proposed explanations for differences in response to treatment include downstream signaling alterations, where resistance had been shown for trastuzumab and pertuzumab but not for T-DM1 [76]. In addition, HER2-positive mouse models have demonstrated that T-DM1 treatment enhances T-cell infiltration compared to trastuzumab, indicating immune-related differences [77].

There are certain limitations within our study. Only patients with HER2-positive breast cancer were included in the cohort; no HER2-negative control group existed. ERBB2 mRNA values may vary depending on the methodology to estimate ERBB2 mRNA levels, which limits comparisons across studies. RT-PCR and in situ hybridization are different methods to evaluate HER2 in tumor material, and some discordant cases were present; for example, some tumors in the low ERBB2 mRNA group had intermediate levels of HER2 amplification group and vice versa. A possible explanation for differences might be that there are tumors with heterogenous HER2 amplification or expression.

Positional scanning (by Cutoff Finder) and STEPP analysis are statistical methods based on mathematical estimations. Before results are confirmed in larger cohorts, the presented cutoffs in this study should not be used as an exact cutoff but rather be regarded as estimations of approximate levels of HER2 copy number, HER2/CEP17 ratio and ERBB2 mRNA where the prognostic value change. The cutoffs differed between DIA algorithm scoring and manual scoring, and this might be explained by the observed difference in the management of HER2 clusters, where DIA algorithm tends to calculate higher HER2 signals in clusters than the manual assessment. The ground truth of cluster estimations would be a DNA sequencing analysis but is outside the scope of this study. Even though the DIA algorithm resulted in different estimations compared to the manual counting, both methods concluded equally that the patients with tumors with the highest and the lowest levels of HER2 gene amplification had a worse prognosis than those with intermediate levels.

Despite limitations, this study presents a unique systematic re-testing of HER2 amplification on whole slide tumor sections. The freshly cut and stained sections provided the best possible quality of slides with the current clinically approved methodology. The reassessments performed by the same resident pathologist and reviewed by one of the two board-certified pathologists blinded to previous HER2 results with the addition of DIA algorithm scoring minimized the risk of variations between observers. DIA algorithm of HER2 copy number and HER2/CEP17 ratio could be a method to standardize assessment and support pathologists’ work, preferably in research where standardization is essential.

Conclusions

In conclusion, the association between either HER2 gene amplification or ERBB2 mRNA and outcome appeared nonlinear. Our results consistently showed that trastuzumab-treated HER2-positive patients with tumors with the highest levels of HER2 gene amplification or ERBB2 mRNA had a worse prognosis than those with intermediate levels. If a future prospective study could show that the patients with tumors with the highest levels of HER2 biomarkers fail on trastuzumab and/or pertuzumab but not T-DXd, this subgroup of patients with the highest levels of HER2 gene amplification or highest levels of ERBB2 mRNA could be offered T-DXd as an initial treatment for early HER2-positive breast cancer. We propose that evaluating biomarker levels parallel to treatment outcomes will be essential to the study design. A first step in understanding HER2 biomarkers and the risk of recurrence is hereby presented. Further studies are ongoing, including molecular studies to understand the biology behind HER2 resistance.

Availability of data and materials

The datasets analyzed during the current study are not publicly available due to local privacy laws but could be available from the corresponding author on reasonable request and after required permissions from the institution and the research collaborators.

Abbreviations

ADC:

Antibody–drug conjugate

AI:

Artificial intelligence

BRISQ:

Biospecimen reporting for improved study quality

CEP17:

Chromosome enumeration probe 17

CI:

Confidence intervals

CYFIP1:

Cytoplasmic FMR1-Interacting Protein 1

DIA:

Digital image analysis

ER:

Estrogen receptor

FFPE:

Formalin-fixed paraffin-embedded

H&E:

Hematoxylin and eosin

HER2:

Human epidermal growth factor receptor 2

HR:

Hazard ratio

IHC:

Immunohistochemistry

ISH:

In situ hybridization

KM:

Kaplan–Meier

PR:

Progesterone receptor

PS:

Positional scanning

RFS:

Recurrence-free survival

STEPP:

Subpopulation treatment effect pattern plot

T-DM1:

Trastuzumab emtansine

T-DXd:

Trastuzumab deruxtecan

References

  1. Slamon DJ, Godolphin W, Jones LA, Holt JA, Wong SG, Keith DE, et al. Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer. Science. 1989;244(4905):707–12. https://doi.org/10.1126/science.2470152.

    Article  ADS  CAS  PubMed  Google Scholar 

  2. Rakha EA, Pinder SE, Bartlett JMS, Ibrahim M, Starczynski J, Carder PJ, et al. Updated UK Recommendations for HER2 assessment in breast cancer. J Clin Pathol. 2015;68(2):93–9. https://doi.org/10.1136/jclinpath-2014-202571.

    Article  PubMed  Google Scholar 

  3. Cardoso F, Kyriakides S, Ohno S, Penault-Llorca F, Poortmans P, Rubio IT, et al. Early breast cancer: ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2019;30(8):1194–220. https://doi.org/10.1093/annonc/mdz173.

    Article  CAS  PubMed  Google Scholar 

  4. Wolff AC, Hammond MEH, Allison KH, Harvey BE, Mangu PB, Bartlett JMS, 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(20):2105–22. https://doi.org/10.1200/JCO.2018.77.8738.

    Article  CAS  PubMed  Google Scholar 

  5. Moelans CB, de Weger RA, Van der Wall E, van Diest PJ. Current technologies for HER2 testing in breast cancer. Crit Rev Oncol Hematol. 2011;80(3):380–92. https://doi.org/10.1016/j.critrevonc.2010.12.005.

    Article  CAS  PubMed  Google Scholar 

  6. 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(11):783–92. https://doi.org/10.1056/nejm200103153441101.

    Article  CAS  PubMed  Google Scholar 

  7. von Minckwitz G, Huang CS, Mano MS, Loibl S, Mamounas EP, Untch M, et al. Trastuzumab emtansine for residual invasive HER2-positive breast cancer. N Engl J Med. 2019;380(7):617–28. https://doi.org/10.1056/NEJMoa1814017.

    Article  Google Scholar 

  8. Baselga J, Cortés J, Kim SB, Im SA, Hegg R, Im YH, et al. Pertuzumab plus trastuzumab plus docetaxel for metastatic breast cancer. N Engl J Med. 2012;366(2):109–19. https://doi.org/10.1056/NEJMoa1113216.

    Article  CAS  PubMed  Google Scholar 

  9. Romond EH, Perez EA, Bryant J, Suman VJ, Geyer CE Jr, Davidson NE, et al. Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. N Engl J Med. 2005;353(16):1673–84. https://doi.org/10.1056/NEJMoa052122.

    Article  CAS  PubMed  Google Scholar 

  10. Krop IE, Beeram M, Modi S, Jones SF, Holden SN, Yu W, et al. Phase I study of trastuzumab-DM1, an HER2 antibody-drug conjugate, given every 3 weeks to patients with HER2-positive metastatic breast cancer. J Clin Oncol. 2010;28(16):2698–704. https://doi.org/10.1200/JCO.2009.26.2071.

    Article  CAS  PubMed  Google Scholar 

  11. Swain SM, Shastry M, Hamilton E. Targeting HER2-positive breast cancer: advances and future directions. Nat Rev Drug Discov. 2023;22(2):101–26. https://doi.org/10.1038/s41573-022-00579-0.

    Article  CAS  PubMed  Google Scholar 

  12. Gianni L, Pienkowski T, Im YH, Roman L, Tseng LM, Liu MC, et al. Efficacy and safety of neoadjuvant pertuzumab and trastuzumab in women with locally advanced, inflammatory, or early HER2-positive breast cancer (NeoSphere): a randomised multicentre, open-label, phase 2 trial. Lancet Oncol. 2012;13(1):25–32. https://doi.org/10.1016/S1470-2045(11)70336-9.

    Article  CAS  PubMed  Google Scholar 

  13. Burstein HJ, Curigliano G, Loibl S, Dubsky P, Gnant M, Poortmans P, et al. Estimating the benefits of therapy for early-stage breast cancer: the St. Gallen International Consensus Guidelines for the primary therapy of early breast cancer 2019. Ann Oncol. 2019;30(10):1541–57. https://doi.org/10.1093/annonc/mdz235.

    Article  CAS  PubMed  Google Scholar 

  14. Lewis Phillips GD, Li G, Dugger DL, Crocker LM, Parsons KL, Mai E, et al. Targeting HER2-positive breast cancer with trastuzumab-DM1, an antibody-cytotoxic drug conjugate. Cancer Res. 2008;68(22):9280–90. https://doi.org/10.1158/0008-5472.CAN-08-1776.

    Article  CAS  PubMed  Google Scholar 

  15. Hurvitz SA, Hegg R, Chung WP, Im SA, Jacot W, Ganju V, et al. Trastuzumab deruxtecan versus trastuzumab emtansine in patients with HER2-positive metastatic breast cancer: updated results from DESTINY-Breast03, a randomised, open-label, phase 3 trial. Lancet. 2023;401(10371):105–17. https://doi.org/10.1016/S0140-6736(22)02420-5.

    Article  CAS  PubMed  Google Scholar 

  16. Cortés J, Kim SB, Chung WP, Im SA, Park YH, Hegg R, et al. Trastuzumab deruxtecan versus trastuzumab emtansine for breast cancer. N Engl J Med. 2022;386(12):1143–54. https://doi.org/10.1056/NEJMoa2115022.

    Article  PubMed  Google Scholar 

  17. 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(1):9–20. https://doi.org/10.1056/NEJMoa2203690.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. van Dooijeweert C, Deckers IAG, Baas IO, van der Wall E, van Diest PJ. Hormone- and HER2-receptor assessment in 33,046 breast cancer patients: a nationwide comparison of positivity rates between pathology laboratories in the Netherlands. Breast Cancer Res Treat. 2019;175(2):487–97. https://doi.org/10.1007/s10549-019-05180-5.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Rüschoff J, Lebeau A, Kreipe H, Sinn P, Gerharz CD, Koch W, et al. Assessing HER2 testing quality in breast cancer: variables that influence HER2 positivity rate from a large, multicenter, observational study in Germany. Mod Pathol. 2017;30(2):217–26. https://doi.org/10.1038/modpathol.2016.164.

    Article  CAS  PubMed  Google Scholar 

  20. Acs B, Fredriksson I, Rönnlund C, Hagerling C, Ehinger A, Kovács A, et al. Variability in breast cancer biomarker assessment and the effect on oncological treatment decisions: a nationwide 5-year population-based study. Cancers. 2021;13(5):1166. https://doi.org/10.3390/cancers13051166.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Yim K, Park HS, Kim DM, Lee YS, Lee A. Image analysis of HER2 immunohistochemical staining of surgical breast cancer specimens. Yonsei Med J. 2019;60(2):158–62. https://doi.org/10.3349/ymj.2019.60.2.158.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Yousif M, Huang Y, Sciallis A, Kleer CG, Pang J, Smola B, et al. Quantitative image analysis as an adjunct to manual scoring of ER, PgR, and HER2 in invasive breast carcinoma. Am J Clin Pathol. 2022;157(6):899–907. https://doi.org/10.1093/ajcp/aqab206.

    Article  CAS  PubMed  Google Scholar 

  23. Li AC, Zhao J, Zhao C, Ma Z, Hartage R, Zhang Y, et al. Quantitative digital imaging analysis of HER2 immunohistochemistry predicts the response to anti-HER2 neoadjuvant chemotherapy in HER2-positive breast carcinoma. Breast Cancer Res Treat. 2020;180(2):321–9. https://doi.org/10.1007/s10549-020-05546-0.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Qaiser T, Mukherjee A, Reddy PBC, Munugoti SD, Tallam V, Pitkäaho T, et al. HER2 challenge contest: a detailed assessment of automated HER2 scoring algorithms in whole slide images of breast cancer tissues. Histopathology. 2018;72(2):227–38. https://doi.org/10.1111/his.13333.

    Article  PubMed  Google Scholar 

  25. La Barbera D, Polónia A, Roitero K, Conde-Sousa E, Della Mea V. Detection of HER2 from haematoxylin-eosin slides through a cascade of deep learning classifiers via multi-instance learning. J Imaging. 2020. https://doi.org/10.3390/jimaging6090082.

    Article  PubMed  PubMed Central  Google Scholar 

  26. Yoder A, Inge LJ, Chen CC, Marati VR, Nguyen TK, Zuiderveld K, et al. Computer-aided scoring of erb-b2 receptor tyrosine kinase 2 (HER2) gene amplification status in breast cancer. J Pathol Inform. 2022;13: 100116. https://doi.org/10.1016/j.jpi.2022.100116.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Gough M, Liu C, Srinivasan B, Wilkinson L, Dunk L, Yang Y, et al. Improved concordance of challenging human epidermal growth factor receptor 2 dual in-situ hybridisation cases with the use of a digital image analysis algorithm in breast cancer. Histopathology. 2023. https://doi.org/10.1111/his.15000.

    Article  PubMed  Google Scholar 

  28. Palm C, Connolly CE, Masser R, Padberg Sgier B, Karamitopoulou E, Simon Q, et al. Determining HER2 status by artificial intelligence: an investigation of primary, metastatic, and HER2 low breast tumors. Diagnostics (Basel). 2023. https://doi.org/10.3390/diagnostics13010168.

    Article  PubMed  Google Scholar 

  29. Tolaney SM, Garrett-Mayer E, White J, Blinder VS, Foster JC, Amiri-Kordestani L, et al. Updated standardized definitions for efficacy end points (STEEP) in adjuvant breast cancer clinical trials: STEEP Version 2.0. J Clin Oncol. 2021;39(24):2720–31. https://doi.org/10.1200/JCO.20.03613.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Sauerbrei W, Taube SE, McShane LM, Cavenagh MM, Altman DG. Reporting recommendations for tumor marker prognostic studies (REMARK): an abridged explanation and elaboration. J Natl Cancer Inst. 2018;110(8):803–11. https://doi.org/10.1093/jnci/djy088.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Moore HM, Kelly AB, Jewell SD, McShane LM, Clark DP, Greenspan R, et al. Biospecimen reporting for improved study quality (BRISQ). Cancer Cytopathol. 2011;119(2):92–101. https://doi.org/10.1002/cncy.20147.

    Article  PubMed  Google Scholar 

  32. American Society of Clinical Oncology/College of American Pathologists Clinical Practice Guideline Focused Update, HER2 testing algorithms [Internet], Alexandria, VA, USA: American Society of Clinical Oncology; 2018. https://old-prod.asco.org/sites/new-www.asco.org/files/content-files/practice-and-guidelines/documents/2018-her2-testing-algorithms.pdf. Accessed 20 Oct 2023.

  33. Hartman J, Ehinger A, Kovács A, Olofsson H, Colon-Cervantes E, Stemme S and Johansson S. Kvalitetsbilaga för bröstpatologi (KVAST-bilaga), [Internet], Sweden: Regionala Cancercentrum i samverkan; 2022 [latest edition updated 2022–02–17]. https://kunskapsbanken.cancercentrum.se/diagnoser/brostcancer/vardprogram/kvalitetsdokument-for--patologi/. Accessed 17 Oct 2023.

  34. Cepheid. Xpert Breast Cancer STRAT4® Assay, [Internet], Sunnyvale, CA, USA: Cepheid; 2023. https://www.cepheid.com/en/tests/Oncology-Genetics/Xpert-Breast-Cancer-STRAT4. Accessed 20 Oct 2023.

  35. Wasserman BE, Carvajal-Hausdorf DE, Ho K, Wong W, Wu N, Chu VC, et al. High concordance of a closed-system, RT-qPCR breast cancer assay for HER2 mRNA, compared to clinically determined immunohistochemistry, fluorescence in situ hybridization, and quantitative immunofluorescence. Lab Investig. 2017;97(12):1521–6. https://doi.org/10.1038/labinvest.2017.93.

    Article  CAS  PubMed  Google Scholar 

  36. Wu NC, Wong W, Ho KE, Chu VC, Rizo A, Davenport S, et al. Comparison of central laboratory assessments of ER, PR, HER2, and Ki67 by IHC/FISH and the corresponding mRNAs (ESR1, PGR, ERBB2, and MKi67) by RT-qPCR on an automated, broadly deployed diagnostic platform. Breast Cancer Res Treat. 2018;172(2):327–38. https://doi.org/10.1007/s10549-018-4889-5.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Budczies J, Klauschen F, Sinn BV, Győrffy B, Schmitt WD, Darb-Esfahani S, et al. Cutoff Finder: a comprehensive and straightforward Web application enabling rapid biomarker cutoff optimization. PLoS ONE. 2012;7(12): e51862. https://doi.org/10.1371/journal.pone.0051862.

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  38. Bates M, Sperinde J, Köstler WJ, Ali SM, Leitzel K, Fuchs EM, et al. Identification of a subpopulation of metastatic breast cancer patients with very high HER2 expression levels and possible resistance to trastuzumab. Ann Oncol. 2011;22(9):2014–20. https://doi.org/10.1093/annonc/mdq706.

    Article  CAS  PubMed  Google Scholar 

  39. Bonetti M, Gelber RD. A graphical method to assess treatment-covariate interactions using the Cox model on subsets of the data. Stat Med. 2000;19(19):2595–609. https://doi.org/10.1002/1097-0258(20001015)19:19%3c2595::aid-sim562%3e3.0.co;2-m.

    Article  CAS  PubMed  Google Scholar 

  40. Bonetti M, Zahrieh D, Cole BF, Gelber RD. A small sample study of the STEPP approach to assessing treatment-covariate interactions in survival data. Stat Med. 2009;28(8):1255–68. https://doi.org/10.1002/sim.3524.

    Article  MathSciNet  PubMed  PubMed Central  Google Scholar 

  41. Lazar AA, Cole BF, Bonetti M, Gelber RD. Evaluation of treatment-effect heterogeneity using biomarkers measured on a continuous scale: subpopulation treatment effect pattern plot. J Clin Oncol. 2010;28(29):4539–44. https://doi.org/10.1200/JCO.2009.27.9182.

    Article  PubMed  PubMed Central  Google Scholar 

  42. Altman DG, De Stavola BL, Love SB, Stepniewska KA. Review of survival analyses published in cancer journals. Br J Cancer. 1995;72(2):511–8. https://doi.org/10.1038/bjc.1995.364.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Therneau TM, Lumley T. Package, “survival.” R Top Doc. 2015;128(10):28–33.

    Google Scholar 

  44. Gelber RD, Wang XV, Cole BF, Cameron D, Cardoso F, Tjan-Heijnen V, et al. Six-year absolute invasive disease-free survival benefit of adding adjuvant pertuzumab to trastuzumab and chemotherapy for patients with early HER2-positive breast cancer: a subpopulation treatment effect pattern plot (STEPP) analysis of the APHINITY (BIG 4–11) trial. Eur J Cancer. 2022;166:219–28. https://doi.org/10.1016/j.ejca.2022.01.031.

    Article  CAS  PubMed  Google Scholar 

  45. Joensuu H, Sperinde J, Leinonen M, Huang W, Weidler J, Bono P, et al. Very high quantitative tumor HER2 content and outcome in early breast cancer. Ann Oncol. 2011;22(9):2007–13. https://doi.org/10.1093/annonc/mdq710.

    Article  CAS  PubMed  Google Scholar 

  46. 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(8):2110–20. https://doi.org/10.1038/s41591-023-02478-2.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  47. Griguolo G, Brasó-Maristany F, González-Farré B, Pascual T, Chic N, Saurí T, et al. ERBB2 mRNA expression and response to ado-trastuzumab emtansine (T-DM1) in HER2-positive breast cancer. Cancers. 2020. https://doi.org/10.3390/cancers12071902.

    Article  PubMed  PubMed Central  Google Scholar 

  48. Perez EA, Hurvitz SA, Amler LC, Mundt KE, Ng V, Guardino E, et al. Relationship between HER2 expression and efficacy with first-line trastuzumab emtansine compared with trastuzumab plus docetaxel in TDM4450g: a randomized phase II study of patients with previously untreated HER2-positive metastatic breast cancer. Breast Cancer Res. 2014;16(3):R50. https://doi.org/10.1186/bcr3661.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  49. Kim SB, Wildiers H, Krop IE, Smitt M, Yu R, de Haas SL, et al. Relationship between tumor biomarkers and efficacy in TH3RESA, a phase III study of trastuzumab emtansine (T-DM1)vs. treatment of physician’s choice in previously treated HER2-positive advanced breast cancer. Int J Cancer. 2016;139(10):2336–42. https://doi.org/10.1002/ijc.30276.

    Article  CAS  PubMed  Google Scholar 

  50. Denkert C, Lambertini C, Fasching PA, Pogue-Geile KL, Mano MS, Untch M, et al. Biomarker data from the phase III KATHERINE study of adjuvant T-DM1 versus trastuzumab for residual invasive disease after neoadjuvant therapy for HER2-positive breast cancer. Clin Cancer Res. 2023;29(8):1569–81. https://doi.org/10.1158/1078-0432.CCR-22-1989.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  51. Xuan Q, Ji H, Tao X, Xu Y, Zhang Q. Quantitative assessment of HER2 amplification in HER2-positive breast cancer: its association with clinical outcomes. Breast Cancer Res Treat. 2015;150(3):581–8. https://doi.org/10.1007/s10549-015-3334-2.

    Article  CAS  PubMed  Google Scholar 

  52. Staaf J, Jönsson G, Ringnér M, Vallon-Christersson J, Grabau D, Arason A, et al. High-resolution genomic and expression analyses of copy number alterations in HER2-amplified breast cancer. Breast Cancer Res. 2010;12(3):R25. https://doi.org/10.1186/bcr2568.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  53. Singer CF, Tan YY, Fitzal F, Steger GG, Egle D, Reiner A, et al. Pathological complete response to neoadjuvant trastuzumab is dependent on HER2/CEP17 ratio in HER2-amplified early breast cancer. Clin Cancer Res. 2017;23(14):3676–83. https://doi.org/10.1158/1078-0432.CCR-16-2373.

    Article  CAS  PubMed  Google Scholar 

  54. Carey LA, Berry DA, Cirrincione CT, Barry WT, Pitcher BN, Harris LN, et al. Molecular heterogeneity and response to neoadjuvant human epidermal growth factor receptor 2 targeting in CALGB 40601, a randomized phase III trial of paclitaxel plus trastuzumab with or without lapatinib. J Clin Oncol. 2016;34(6):542–9. https://doi.org/10.1200/JCO.2015.62.1268.

    Article  CAS  PubMed  Google Scholar 

  55. Scaltriti M, Nuciforo P, Bradbury I, Sperinde J, Agbor-Tarh D, Campbell C, et al. High HER2 expression correlates with response to the combination of lapatinib and trastuzumab. Clin Cancer Res. 2015;21(3):569–76. https://doi.org/10.1158/1078-0432.CCR-14-1824.

    Article  CAS  PubMed  Google Scholar 

  56. Fumagalli D, Venet D, Ignatiadis M, Azim HA Jr, Maetens M, Rothé F, et al. RNA sequencing to predict response to neoadjuvant anti-HER2 therapy: a secondary analysis of the NeoALTTO randomized clinical trial. JAMA Oncol. 2017;3(2):227–34. https://doi.org/10.1001/jamaoncol.2016.3824.

    Article  PubMed  PubMed Central  Google Scholar 

  57. Dowsett M, Procter M, McCaskill-Stevens W, de Azambuja E, Dafni U, Rueschoff J, et al. Disease-free survival according to degree of HER2 amplification for patients treated with adjuvant chemotherapy with or without 1 year of trastuzumab: the HERA trial. J Clin Oncol. 2009;27(18):2962–9. https://doi.org/10.1200/jco.2008.19.7939.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  58. Mercogliano MF, De Martino M, Venturutti L, Rivas MA, Proietti CJ, Inurrigarro G, et al. TNFα-induced mucin 4 expression elicits trastuzumab resistance in HER2-positive breast cancer. Clin Cancer Res. 2017;23(3):636–48. https://doi.org/10.1158/1078-0432.CCR-16-0970.

    Article  CAS  PubMed  Google Scholar 

  59. Scaltriti M, Rojo F, Ocaña A, Anido J, Guzman M, Cortes J, et al. Expression of p95HER2, a truncated form of the HER2 receptor, and response to anti-HER2 therapies in breast cancer. J Natl Cancer Inst. 2007;99(8):628–38. https://doi.org/10.1093/jnci/djk134.

    Article  CAS  PubMed  Google Scholar 

  60. Nagy P, Friedländer E, Tanner M, Kapanen AI, Carraway KL, Isola J, et al. Decreased accessibility and lack of activation of ErbB2 in JIMT-1, a herceptin-resistant, MUC4-expressing breast cancer cell line. Cancer Res. 2005;65(2):473–82.

    Article  CAS  PubMed  Google Scholar 

  61. Agus DB, Akita RW, Fox WD, Lewis GD, Higgins B, Pisacane PI, et al. Targeting ligand-activated ErbB2 signaling inhibits breast and prostate tumor growth. Cancer Cell. 2002;2(2):127–37. https://doi.org/10.1016/s1535-6108(02)00097-1.

    Article  CAS  PubMed  Google Scholar 

  62. Agus DB, Gordon MS, Taylor C, Natale RB, Karlan B, Mendelson DS, et al. Phase I clinical study of pertuzumab, a novel HER dimerization inhibitor, in patients with advanced cancer. J Clin Oncol. 2005;23(11):2534–43. https://doi.org/10.1200/JCO.2005.03.184.

    Article  CAS  PubMed  Google Scholar 

  63. Loibl S, von Minckwitz G, Schneeweiss A, Paepke S, Lehmann A, Rezai M, et al. PIK3CA mutations are associated with lower rates of pathologic complete response to anti-human epidermal growth factor receptor 2 (HER2) therapy in primary HER2-overexpressing breast cancer. J Clin Oncol. 2014;32(29):3212–20. https://doi.org/10.1200/jco.2014.55.7876.

    Article  CAS  PubMed  Google Scholar 

  64. Park S, Jiang Z, Mortenson ED, Deng L, Radkevich-Brown O, Yang X, et al. The therapeutic effect of anti-HER2/neu antibody depends on both innate and adaptive immunity. Cancer Cell. 2010;18(2):160–70. https://doi.org/10.1016/j.ccr.2010.06.014.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  65. Oliveras-Ferraros C, Corominas-Faja B, Cufí S, Vazquez-Martin A, Martin-Castillo B, Iglesias JM, et al. Epithelial-to-mesenchymal transition (EMT) confers primary resistance to trastuzumab (Herceptin). Cell Cycle. 2012;11(21):4020–32. https://doi.org/10.4161/cc.22225.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  66. Liu H, Lyu H, Jiang G, Chen D, Ruan S, Liu S, et al. ALKBH5-mediated m6A demethylation of GLUT4 mRNA promotes glycolysis and resistance to HER2-targeted therapy in breast cancer. Cancer Res. 2022;82(21):3974–86. https://doi.org/10.1158/0008-5472.CAN-22-0800.

    Article  CAS  PubMed  Google Scholar 

  67. Hori A, Shimoda M, Naoi Y, Kagara N, Tanei T, Miyake T, et al. Vasculogenic mimicry is associated with trastuzumab resistance of HER2-positive breast cancer. Breast Cancer Res. 2019;21(1):88. https://doi.org/10.1186/s13058-019-1167-3.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  68. Filho OM, Viale G, Stein S, Trippa L, Yardley DA, Mayer IA, et al. Impact of HER2 heterogeneity on treatment response of early-stage HER2-positive breast cancer: phase II neoadjuvant clinical trial of t-dm1 combined with pertuzumab. Cancer Discov. 2021;11(10):2474–87. https://doi.org/10.1158/2159-8290.cd-20-1557.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  69. Puig T, Aguilar H, Cufí S, Oliveras G, Turrado C, Ortega-Gutiérrez S, et al. A novel inhibitor of fatty acid synthase shows activity against HER2+ breast cancer xenografts and is active in anti-HER2 drug-resistant cell lines. Breast Cancer Res. 2011;13(6):R131. https://doi.org/10.1186/bcr3077.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  70. Ram S, Kim D, Ober RJ, Ward ES. The level of HER2 expression is a predictor of antibody-HER2 trafficking behavior in cancer cells. MAbs. 2014;6(5):1211–9. https://doi.org/10.4161/mabs.29865.

    Article  PubMed  PubMed Central  Google Scholar 

  71. Chang MC, Malowany JI, Mazurkiewicz J, Wood M. “Genetic heterogeneity” in HER2/neu testing by fluorescence in situ hybridization: a study of 2522 cases. Mod Pathol. 2012;25(5):683–8. https://doi.org/10.1038/modpathol.2011.206.

    Article  CAS  PubMed  Google Scholar 

  72. Allison KH, Dintzis SM, Schmidt RA. Frequency of HER2 heterogeneity by fluorescence in situ hybridization according to CAP expert panel recommendations: time for a new look at how to report heterogeneity. Am J Clin Pathol. 2011;136(6):864–71. https://doi.org/10.1309/AJCPXTZSKBRIP07W.

    Article  PubMed  Google Scholar 

  73. Buckley NE, Forde C, McArt DG, Boyle DP, Mullan PB, James JA, et al. Quantification of HER2 heterogeneity in breast cancer-implications for identification of sub-dominant clones for personalised treatment. Sci Rep. 2016;6:23383. https://doi.org/10.1038/srep23383.

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  74. Hunter FW, Barker HR, Lipert B, Rothé F, Gebhart G, Piccart-Gebhart MJ, et al. Mechanisms of resistance to trastuzumab emtansine (T-DM1) in HER2-positive breast cancer. Br J Cancer. 2020;122(5):603–12. https://doi.org/10.1038/s41416-019-0635-y.

    Article  CAS  PubMed  Google Scholar 

  75. Junttila TT, Li G, Parsons K, Phillips GL, Sliwkowski MX. Trastuzumab-DM1 (T-DM1) retains all the mechanisms of action of trastuzumab and efficiently inhibits growth of lapatinib insensitive breast cancer. Breast Cancer Res Treat. 2011;128(2):347–56. https://doi.org/10.1007/s10549-010-1090-x.

    Article  CAS  PubMed  Google Scholar 

  76. Ferraro E, Safonov A, Chen Y, White C, Marra A, Ahmed M, et al. Abstract P4–02–01: efficacy of HER2 ADCs against HER2 inhibitor resistance alterations in the PI3K and MAPK pathways in HER2-positive breast cancer. Cancer Res. 2023;83(5):P4-02–01. https://doi.org/10.1158/1538-7445.SABCS22-P4-02-01.

    Article  Google Scholar 

  77. Müller P, Kreuzaler M, Khan T, Thommen DS, Martin K, Glatz K, et al. Trastuzumab emtansine (T-DM1) renders HER2+ breast cancer highly susceptible to CTLA-4/PD-1 blockade. Sci Transl Med. 2015;7(315):315. https://doi.org/10.1126/scitranslmed.aac4925.

    Article  CAS  Google Scholar 

Download references

Acknowledgements

We thank Landon Inge, Petra De Verdier, Simon Weichelt and Lina Lundberg at Roche for excellent collaboration, valuable input on the uPath HER2 Dual ISH image analysis, breast algorithm, results and manuscript. We thank Lisa Viberg and Ran Ma for their support with the Xpert Breast Cancer STRAT4® Assay and their excellent research assistance within the Hartman research group at Karolinska Institutet. In addition, we thank Marine Gergondey at Cepheid for her excellent collaboration and valuable input regarding the Xpert Breast Cancer STRAT4® Assay. We thank Ulla Wilking at the Department of Oncology and Pathology, Karolinska Institutet, for initiating a study that involved HER2-positive breast cancer patients. Moreover, we thank the Pathology Core Facility at Karolinska University Laboratory for providing laboratory services.

Funding

Open access funding provided by Karolinska Institute. This study was supported by grants from Region Stockholm Sweden, the Swedish Cancer Society, MedTechLabs, the Cancer Society in Stockholm, the Swedish Breast Cancer Association and the Swedish Society for Medical Research. Parts of this study were performed as a research collaboration between Karolinska Institutet and Cepheid. Cepheid contributed to this study through financial support and material for STRAT4 ERBB2 mRNA analysis and provided input on statistical analysis and the manuscript. Parts of this study were performed as a research collaboration between Karolinska Institutet and Roche. Roche contributed to this study by providing access to the VENTANA DP 200 slide scanner, the Roche uPath enterprise software, the uPath HER2 Dual ISH image analysis, breast algorithm and input on the manuscript.

Author information

Author notes

  1. Stephanie Robertson and Johan Hartman shared last authorship—equal contributions.

Authors and Affiliations

  1. Department of Oncology and Pathology, Karolinska Institutet, Visionsgatan 56, CCK R8:04, 17176, Stockholm, Sweden

    Caroline Rönnlund, Emmanouil G. Sifakis, Caroline Schagerholm, Qiao Yang, Emelie Karlsson, Xinsong Chen, Theodoros Foukakis, Stephanie Robertson & Johan Hartman

  2. Department of Clinical Pathology and Cancer Diagnostics, Karolinska University Hospital, Stockholm, Sweden

    Caroline Rönnlund & Johan Hartman

  3. Breast Center, Theme Cancer, Karolinska University Hospital, Stockholm, Sweden

    Theodoros Foukakis

  4. Medical and Scientific Affairs and Strategy, Oncology, Cepheid, Sunnyvale, CA, USA

    Jodi Weidler & Michael Bates

  5. Department of Molecular Medicine and Surgery, Karolinska Institutet, Stockholm, Sweden

    Irma Fredriksson

  6. Department of Breast-, Endocrine Tumors and Sarcoma, Karolinska University Hospital, Stockholm, Sweden

    Irma Fredriksson

  7. Medtechlabs, Bioclinicum, Karolinska University Hospital, Stockholm, Sweden

    Johan Hartman

Authors
  1. Caroline Rönnlund
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Emmanouil G. Sifakis
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Caroline Schagerholm
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Qiao Yang
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Emelie Karlsson
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Xinsong Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Theodoros Foukakis
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Jodi Weidler
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Michael Bates
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Irma Fredriksson
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Stephanie Robertson
    View author publications

    You can also search for this author in PubMed Google Scholar

  12. Johan Hartman
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

CR, TF, IF, SR and JH performed study concept and design; CR and SR performed data collection; CS performed selected data collection; XC and JW supported with Xpert Breast Cancer STRAT4® Assay; EK performed research assistance and scanning of the tumor slides; CR performed manual HER2 IHC and HER2 ISH analysis and DIA HER2 ISH analysis; SR and JH reviewed manual HER2 IHC and ISH analysis; CR performed statistical analysis; EGS wrote the code and supervised the bioinformatics/ statistical analysis; QY and SR performed selected statistical analysis; CR, EGS, CS, QY, EK, XC, TF, JW, MB, IF, SR and JH performed interpretation of data; CR wrote the manuscript draft and arranged visualization; SR wrote selected parts of the manuscript draft; C.R., EGS, CS, QY, EK, XC, TF, JW, MB, IF, SR and JH performed manuscript editing and manuscript review; SR and JH supervised the study; JH provided resources. All authors have read and approved the final paper.

Corresponding author

Correspondence to Caroline Rönnlund.

Ethics declarations

Ethics approval and consent to participate

This study was conducted according to the guidelines of the Declaration of Helsinki and was approved by the Regional Ethical Review Board in Stockholm. No additional informed consent was required in accordance with ethical approval in this non-interventional data collection and analysis.

Consent for publication

Not applicable.

Competing interests

J.H. has obtained speaker's honoraria or advisory board remunerations from Roche, Novartis, AstraZeneca, Pfizer, Eli Lilly, MSD and Gilead and has received institutional research support from Cepheid, Roche, Novartis and AstraZeneca. J.H. is a co-founder and shareholder of Stratipath AB. T.F. has obtained speaker's honoraria or advisory board remunerations from Astra Zeneca, Roche, Gilead, Affibody, Pfizer, Exact Sciences, Novartis and Veracyte and has received institutional research support from Astra Zeneca, Novartis, Roche and Pfizer and author royalties from UpToDate. I.F. has received institutional research support from MSD. S.R. is an employee and holds stock options of Stratipath AB. J.W. and M.B. are full-time employees of Cepheid and own stock in Danaher, Inc. All other authors have no disclosures to declare.

Additional information

Publisher's Note

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

Supplementary Information

Additional file 1

:Table S1. BRISQ criteria. Table S2. Tumor characteristics of the Stockholm HER2 cohort. Table S3. Re-tested HER2 immunohistochemistry and in situ hybridization (ISH) results. Table S4. Adjuvant treatment data of the Stockholm HER2 cohort. Table S5. Outcome data of the Stockholm HER2 cohort. Table S6. Suggested biomarker groups by Cutoff Finder and STEPP functions.

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/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Rönnlund, C., Sifakis, E.G., Schagerholm, C. et al. Prognostic impact of HER2 biomarker levels in trastuzumab-treated early HER2-positive breast cancer. Breast Cancer Res 26, 24 (2024). https://doi.org/10.1186/s13058-024-01779-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13058-024-01779-9

Keywords

Breast Cancer Research

ISSN: 1465-542X

Contact us