PDF
Citation
Print
Introduction
Breast cancer is one of the most prevalent cancers and the leading cause of cancer-related death among women globally, imposing significant social and economic burdens [1]. This complex and heterogeneous disease has various molecular subtypes that dictate treatment strategies and clinical outcomes. Among these subtypes, human epidermal growth factor receptor 2 (HER2) has been considered a critical biomarker in guiding therapeutic decisions. HER2 status is currently determined using the American Society of Clinical Oncology/College of American Pathologists (ASCO/CAP) guidelines. HER2-overexpressing tumors based on immunohistochemistry (IHC) 3+ or IHC 2+ with confirmation of HER2 gene amplification through in situ hybridization (ISH) positivity are considered “HER2-positive,” while tumors with IHC 0, 1+, or 2+ with ISH negativity are considered “HER2-negative.” HER2-positive breast cancer accounts for approximately 15%-30% of breast cancer cases and these patients can benefit from anti-HER2 targeting agents, such as monoclonal antibodies [mAbs] (trastuzumab, pertuzumab), tyrosine kinase inhibitors (tucatinib, neratinib, lapatinib), or antibody-drug conjugates [ADCs] (trastuzumab-emtansine [T-DM1], trastuzumab-deruxtecan [T-DXd]). However, HER2-negative tumors have not traditionally been considered eligible candidates for HER2-targeting agents because of the absence of oncogene addiction from HER2 gene amplification.
HER2-low breast cancer, defined as a tumor with an IHC score of 1+ or 2+ and ISH negativity, has emerged as an intriguing entity that challenges our understanding of HER2-driven malignancies. Notably, the DESTINY-Breast04 (DB-04) clinical trial has demonstrated the remarkable efficacy of T-DXd compared with physicians’ choice chemotherapy in patients with previously treated metastatic HER2-low breast cancer [2]. These findings have challenged the conventional belief that HER2-low tumors are unresponsive to HER2- targeted therapies. As a result, there is a growing need for further evaluation to enhance our understanding of the biological and clinical characteristics of HER2-low breast cancer and to develop optimal treatment strategies for this specific subtype. Despite the numerous studies conducted over the past several years, several important questions and uncertainties about HER2-low breast cancer remain, demanding further investigation.
This review aims to summarize the current evidence, identify knowledge gaps, and highlight future research directions regarding HER2-low breast cancer. By examining the existing literature, we hope to contribute to the growing understanding of this intriguing subtype of breast cancer and to lay the foundation for optimized diagnostic and therapeutic approaches to ultimately improve patient outcomes in the era of precision oncology.
Characteristics of HER2-Low Tumors
According to the 2018 ASCO/CAP guidelines, HER2- low tumors are defined as tumors with an IHC score of 1+ or 2+ without gene amplification [2]. Based on this definition, what was traditionally categorized as “HER2-negative” breast cancer is now further subdivided into “HER2-low” and “HER2-zero,” the latter being tumors with IHC scores of 0. The critical question is whether HER2-low tumors represent a subtype distinct from HER2-zero tumors at either the molecular or clinical level, potentially indicating a new entity with particular prognostic implications.
1. Clinical characteristics of HER2-low breast cancer
Numerous prior studies have evaluated the characteristics of HER2-low tumors and their prognostic implications compared with HER2-zero tumors [3-11]. Most studies have indicated that HER2-low breast cancer exhibits a higher prevalence of hormone receptor positivity (HR+) [3-5,8,10-12], lower nuclear and histologic grades [4,5,8], and a lower Ki-67 proliferation index [4,5,7,12] than HER2-zero breast cancer. Kang et al. [5] conducted a multivariable logistic regression analysis to elucidate the association between HER2-low and hormone receptor (HR) status and revealed that HR+ had a strong association with HER2-low (odds ratio [OR], 3.1; 95% confidence interval [CI], 2.35 to 4.11; p < 0.001). In contrast, the histologic grade and Ki-67 index did not show a significant association with HER2-low. Furthermore, Tarantino et al. demonstrated that the proportion of HER2-low tumors progressively increased with increasing thresholds of estrogen receptor expression (p < 0.001) [8]. These findings suggest that HR status is significantly related to characteristics of HER2-low breast cancer.
2. Molecular characteristics of HER2-low breast cancer
Several studies have explored the molecular characteristics of HER2-low breast cancer [11,13-17]. Schettini et al. [11] conducted PAM50 subtype and individual gene expression analyses in 13 independent datasets (n=3,689). They demonstrated that luminal-related genes and two PAM50 signatures (i.e., luminal A and B) were upregulated in HER2-low tumors, while proliferation-related genes, basal-like-related genes, tyrosine kinase receptors, and three PAM50 signatures (i.e., HER2-enriched, basal-like, normal-like) were downregulated in these tumors [11]. When separately analyzed according to HR status, similar results to those observed in the entire population were observed in the HR+ groups, whereas no differences were observed in the HR-negative (HR–) subgroup (also known as triple-negative breast cancer [TNBC]) [11]. Notably, most tumors in the HR+ subgroup were the luminal type in the PAM50 analysis, regardless of HER2-IHC expression (86.7% in HER2-IHC0, 91% in HER2-IHC1+, and 94.4% in HER2-IHC2+ without gene amplification). In contrast, most tumors in the TNBC subgroup were the basal-like type, regardless of HER2-IHC expression (85% in HER2-IHC0, 85.4% in HER2-IHC1+, and 78.4% in HER2-IHC2+ without gene amplification).
Agostinetto et al. [13] also conducted a PAM50 subtype analysis using the Total Cancer Genome Atlas (TCGA) dataset and showed that the distribution of intrinsic subtypes among HR+/HER2-low tumors resembled that of the subtypes among HR+/HER2-zero tumors rather than that of the subtypes among HR–/HER2-low (TNBC) tumors. Specifically, most HR+/HER2-low tumors (91.3%) and HR+/HER2-zero tumors (91.1%) were classified as the luminal type. Conversely, among HR–/HER2-low tumors, the most common subtype was the basal-like type (76.7%), followed by HER2-enriched (13.7%), and the luminal A type accounted for only 1.4%. Zhang et al. [14] also demonstrated similar results using Mammaprint and Blueprint results. The authors reported that most HR+/HER2-low tumors were luminal types A and B (70.4% and 28.4%, respectively), while 66.7% of HR–/HER2-low tumors were the basal-like type and 33.3% were the luminal type B. These studies consistently demonstrated that regardless of HER2-expression, the HR+ subgroup is predominantly of the luminal type and that the TNBC subgroup is primarily basal-like on PAM50 analysis. This suggests that intrinsic subtypes of breast cancer are more notably associated with HR status than HER2-low status or its absence.
Jin et al. [15] conducted next-generation sequencing on their metastatic breast cancer (mBC) cohort (n=445) and the TCGA dataset. They found no significant differences in common gene mutations and copy number variations between HER2- low and HER2-zero tumors after correcting for HR status. Similarly, Tarantino et al. [16] performed the most extensive genomic landscape study (n=1,043) in a HER2-negative mBC cohort and confirmed that there were no notable differences in common genetic mutations and tumor mutational burden between HER2-low and HER2-zero tumors after adjusting for estrogen receptor status. These findings were consistent with those of Bansal et al. [18] and Marra et al. [19], indicating HER2-low breast cancer did not have distinct molecular characteristics. In contrast, Berrino et al. [17] showed that somatic mutations of a HER2-low cohort were significantly different from those of a HER2-positive or HER2-negative matched cohort (the Memorial Sloan-Kettering Cancer Center cohort in cBioportal) using DNA and RNA high-throughput analysis (n=99) [17]. Nevertheless, the authors did not conduct a subgroup analysis according to HR status, which limits the interpretation of this study.
To summarize the abovementioned studies, HER2-low breast cancer is not considered a distinct entity at the molecular level compared with HER2-zero breast cancer, and HR status is recognized as the key driver of the biology of HER2-low breast cancer, as shown by clinical and molecular characteristics. The Expert Consensus Statements by the European Society for Medical Oncology (ESMO) suggest that HER2-low breast cancer should not be considered a separate and distinct molecular entity [20]. Instead, it should be recognized as a diverse group of tumors for which the presence or absence of HR expression primarily influences their biological characteristics [20].
3. Prognostic role of HER2-low in breast cancer
Although HER2-low is not considered a distinct entity, whether HER2-low status is an independent prognostic factor remains controversial. The rate of pathological complete response (pCR) of HER2-low breast cancer after neoadjuvant chemotherapy is lower than that of HER2-zero breast cancer in most studies [4,5,8,21-25]; however, after correcting for HR status, no significant differences in pCR rates are observed between the two subsets [4-8,12,21-26] (Table 1). A recent meta-analysis including 42 studies showed similar results: HER2-low tumors had lower rates of pCR than HER2-zero tumors in the entire cohort (OR, 0.74; 95% CI, 0.62 to 0.88; p=0.001) and in the HR+ subset (OR, 0.77; 95% CI, 0.65 to 0.9; p=0.001), whereas no differences were observed in the HR–subset (OR, 0.95; 95% CI, 0.81 to 1.11; p=0.497) [27].
Regarding survival outcomes, there is conflicting evidence of the prognostic role of HER2-low status. Certain studies have indicated that HER2-low breast cancer presents either better [3,4,28,29] or worse [30] clinical outcomes than HER2-zero breast cancer. In contrast, other studies found no significant survival differences between these two groups, especially after correcting for HR status [5-8,10,11,31,32] (Table 2). The same meta-analysis described above revealed that HER2-low status was an independent prognostic factor associated with disease-free survival (DFS) (hazard ratio, 0.86; 95% CI, 0.79 to 0.92; p < 0.001) and overall survival (OS) (hazard ratio, 0.9; 95% CI, 0.85 to 0.95; p < 0.001) in the entire population. When conducting separate analyses according to HR status, similar results were observed within HR+ subset (DFS: hazard ratio, 0.86; 95% CI, 0.80 to 0.93; p < 0.001; OS: hazard ratio, 0.94; 95% CI, 0.9 to 0.98; p=0.003). In contrast, no differences were observed in DFS among the HR– subset in the early setting (p=0.155); however, OS was better in the HER2-low group than the HER2-zero group (hazard ratio, 0.88; 95% CI, 0.82 to 0.95; p=0.001) [27]. In the patients with mBC, HER2-low showed a significant association with better OS across the overall population, regardless of HR status (p=0.008 in the overall population; p=0.013 in the HR+subset; p < 0.001 in the HR– subset). However, no substantial differences in progression-free survival (PFS) between the two subsets were detected in the overall population or within subgroups with different HR statuses (p=0.710 in the overall population; p=0.192 in the HR+ subset; p=0.103 in the HR– subset) [27]. Although the meta-analysis revealed better OS and DFS in the HR+ subset in patients with early breast cancer and better OS in patients with mBC, these differences were marginal and exhibited inconsistency across HR statuses. The present evidence is inadequate to draw a definitive conclusion, highlighting the need for further investigation of the prognostic role of HER2-low status.
4. Changes in HER2-low status during the disease course
Several studies have recently investigated the transition of HER2 status, including the HER2-low concept, in both early and advanced settings [24,33-45]. The discordant rate of HER2 status between the primary tumor and residual disease after neoadjuvant chemotherapy was 21.4%-36.4% [24,33-35,42]. Similarly, when evaluating the HER2 status of the primary tumor and a paired recurrent lesion and/or metastatic lesion, substantial portions of tumors showed a discordant HER2 status, ranging from 27.6% to 38% [36-41,43- 45]. Furthermore, HER2-status changes were also observed between the first biopsy of metastatic lesions and subsequent biopsies during the disease course, and the discordant rate was nearly 50% [40,44]. These findings indicate instability of HER2-low status throughout the disease course, suggesting the need for re-evaluation of HER2-status in cases of residual disease after neoadjuvant chemotherapy and/or in cases of recurrent/metastatic lesions.
Only a few studies have reported the prognostic role of a HER2 transition (HER2-low to HER2-zero or vice versa). Miglietta et al. [24] showed that a HER2 transition was not significantly related to DFS, while the German Breast Group reported that patients with a HER2 transition from HER2-zero to HER2-low showed a significantly reduced invasive DFS compared with patients with a constant HER2-low status [46]. Kang et al. [34] showed that patients with a HER2 transition from HER2-low to HER2-zero had better DFS and OS than patients with a constant HER2-zero status; however, when separately analyzed according to HR status, no significant differences in DFS or OS were observed. Due to limited evidence, currently, the prognostic role of a HER2 transition remains unknown, and further evaluation is necessary.
The mechanisms of a HER2 transition (HER2-low to HER2-zero or vice versa) have not been delineated well, but the possible reasons are as follows: (1) the intratumoral heterogeneity of breast cancer [47-49] and (2) therapeutic selection pressure after systemic treatment (including surgery, chemotherapy, hormonal therapy, and anti-HER2 therapy) [44,50,51]. Notably, patients with HR+/HER2-zero breast cancer have a higher probability of a HER2-low transition than patients with HR–/HER2-zero breast cancer in both early (OR, 2.48; 95% CI, 1.62 to 3.87) [34] and advanced settings (OR, 2.66; 95% CI, 1.12 to 6.59; p=0.027) [38], supporting the strong association between HR+ and HER2-low tumors in biology and that HR+ might affect changes in HER2 status. In conclusion, HER2-low breast cancer is not currently considered a new entity of breast cancer based on the cumulative evidence reported to date. The prognostic role of HER2-low status and its transition after treatment remains controversial.
Treatment of Advanced HER2-Low Tumors
1. T-DXd
Before introducing T-DXd, several studies evaluated the role of anti-HER2-targeting agents in advanced HER2-low breast cancer; however, none demonstrated clinical efficacy. Pertuzumab did not show a meaningful clinical benefit; only six patients showed a partial response (PR) or stable disease over 24 weeks (6/78, 7.6%) [52]. Margetuximab (an antiHER2 mAb) showed in vitro activity against HER2-low cell lines [53], but its clinical efficacy was not confirmed because the phase II trial was prematurely ended due to insufficient efficacy (NCT01828021) [54]. T-DM1 also showed disappointing results among patients with heavily treated metastatic HER2-low breast cancer, with a median PFS of 2.6 months, and the objective response rate (ORR) was 4.8% in a phase II trial [55].
T-DXd is an ADC including trastuzumab and deruxtecan (a topoisomerase I inhibitor) as a payload connected by a cleavable tetrapeptide-based linker [56]. T-DXd has a higher drug-antibody ratio [DAR] than T-DM1 (8 vs. 3.5), and the released payload can easily cross the cell membrane, which potentially has a cytotoxic effect on neighboring tumor cells regardless of target expression as a bystander effect [56,57]. T-DXd has shown clinical benefits in second and subsequent lines in HER2-positive mBC [58-60], and clinical trials are now ongoing to evaluate its efficacy in first-line treatment (DESTINY-Breast09 trial; NCT04784715) and adjuvant settings (DESTINY-Breast05 trial; NCT04622319).
1) Results of the DB-04 trial
The DB-04 trial was a phase 3 study that showed the benefit of T-DXd in HER2-low unresectable/mBC [2]. A total of 557 patients who were previously treated with chemotherapy (or endocrine treatment in the case of HR+) for metastatic disease or had disease recurrence during or within six months after receiving adjuvant chemotherapy were enrolled. Patients included in the study were randomized in a 2:1 ratio to the T-DXd group or the physician’s choice chemotherapy group. In the study population, most patients had HR+ disease (n=494, 88.7%), and only a small proportion had HR– disease (n=63, 11.3%). T-DXd significantly reduced the risk of disease progression and death by approximately 50% (median PFS, 9.9 months vs. 5.1 months; hazard ratio, 0.5; p < 0.001) and 36% (median OS, 23.4 months vs. 16.8 months; hazard ratio, 0.64; p=0.001), respectively. These findings were consistent regardless of HR status, HER2 status (IHC1+ vs. IHC2+without gene amplification), or prior exposure to a cyclindependent kinase 4 and 6 (CDK4/6) inhibitor in the HR+subset. Regarding safety, serious adverse events (AEs) were similar in both groups (27.8% in the T-DXd group, 25.0% in the control group). However, the rate of drug discontinuation due to AEs was higher in the T-DXd group than in the control group (16.2% vs. 8.1%). The most common grade 3 or higher toxicities were neutropenia (13.7%), anemia (8.1%), and fatigue (7.5%) in the T-DXd group. Drug-related interstitial lung disease or pneumonitis occurred in 45 patients (12.1%) in the T-DXd group, including five patients with grade 3 and three with grade 5 toxicity.
Based on the results of the DB-04 trial, which demonstrated significant clinical efficacy and tolerable safety, T-DXd was approved by the United States Food and Drug Administration (U.S. FDA) for unresectable or metastatic HER2-low breast cancer on August 5, 2022.
2) Other studies of the efficacy of T-DXd for HER2-low breast cancer
The phase II DAISY trial (NCT04132960) aimed to evaluate the efficacy of T-DXd according to HER2 expression in mBC. Patients with previously treated mBC were enrolled and categorized according to their HER2 status (their most recent known HER2 status) into cohort 1 (HER2-positive, n=72), 2 (HER2-low, n=74), and 3 (HER2-zero, n=40) [61]. The primary endpoint of this study was the confirmed ORR, and it was met in cohort 1 (70.6%; 95% CI, 58.3 to 81) and cohort 2 (37.5%; 95% CI, 26.4 to 49.7) but not in cohort 3 (29.7%; 95% CI, 15.9 to 47). Consistent with the findings of the DB-04 trial, the DAISY trial also demonstrated the clinical efficacy of T-DXd for HER2-low mBC.
Several studies have evaluated the effectiveness of combining T-DXd and immune checkpoint inhibitors. The phase Ib/II BEGONIA trial investigated the efficacy of T-DXd combined with durvalumab in arm 6. A total of 11 patients were included, and the confirmed ORR was 100% (4/4; only four patients had the opportunity to complete two on-treatment evaluations). Grade 3 or higher toxicities were observed in four patients (36%), and serious AEs were experienced by one patient (9%) [62]. T-DXd combined with nivolumab (360 mg) was also assessed in another phase Ib trial, revealing an ORR of 37.5% (6/16, all PR) and a disease control rate (DCR) of 75% among patients with heavily treated HER2-low mBC. The median PFS was 6.3 months (95% CI, 2.3 to not evaluated), and the safety profiles were similar to those from earlier studies [63].
2. Sacituzumab govitecan
Sacituzumab govitecan (SG) is an ADC targeting trophoblast cell-surface antigen 2 (TROP2). It consists of an anti-TROP2 mAb and SN-38 (a topoisomerase I inhibitor) as a payload connected by a hydrolyzable linker [64-66]. SG showed promising efficacy compared with single-agent chemotherapy among patients with TNBC who were previously treated with at least two lines of chemotherapy in the ASCENT trial (median PFS, 5.6 vs. 1.7 months; hazard ratio, 0.41; p < 0.001; median OS, 12.1 vs. 6.7 months; hazard ratio, 0.48; p < 0.001) [67]. In the post hoc analysis of the ASCENT trial, 26% of patients (n=123) had HER2-low tumors, and 62% (n=293) had HER2-zero tumors [68]. The median PFS and OS were significantly better than those in the chemotherapy group in both the HER2-low and HER2-zero groups (median PFS, 6.2 months vs. 2.9 months for HER2-low, 4.3 months vs. 1.6 months for HER2-zero; median OS, 14.0 months vs. 8.7 months for HER2-low, 11.3 months vs. 5.9 months for HER2- zero), suggesting that the efficacy of SG is consistent regardless of HER2-IHC status.
In addition, the TROPiCS-02 trial investigated the effectiveness of SG among pretreated HR+/HER2-negative breast cancer [69]. Patients with metastatic or unresectable HR+/HER2-negative tumors that progressed after at least one endocrine therapy, taxane, and CDK4/6 inhibitor in any setting (n=543) were randomized 1:1 to an SG group and a physician’s choice chemotherapy group. SG also demonstrated a significant reduction in the risk of progression or death (median PFS, 5.5 vs. 4.0 months; HR, 0.65; p < 0.001) and improved OS (median OS, 14.5 vs. 11.2 months; HR, 0.79; p=0.01) compared with the physician’s choice chemotherapy. Regarding AEs, the most common grade 3 or higher treatment-related AEs (> 5% incidence) in the SG group were neutropenia, leukopenia, diarrhea, anemia, and fatigue. Serious treatment-related AEs were reported in 37 patients (14%) in the SG group, and the most common AEs were diarrhea (5%), febrile neutropenia (4%), neutropenia (3%), and neutropenic colitis (2%). Six patients died due to AEs, including one patient due to treatment-related AEs.
Recently, post hoc analysis results from the TROPiCS-02 trial were presented to evaluate the efficacy of SG according to HER2-IHC status [70,71]. HER2 status was determined by local IHC assessment on the last available pathology sample, and 57% of patients had HER2-low tumors, while 43% had HER2-zero tumors. In terms of PFS, SG was associated with a better PFS than the control arm in both the HER2-low (median PFS, 6.4 months vs. 4.2 months; hazard ratio, 0.58; p < 0.001) and HER2-zero subgroups (median PFS, 5.0 months vs. 3.4 months; hazard ratio, 0.72; p < 0.001) [70]. Similarly, SG demonstrated a reduced risk of death in both subgroups compared with the control arm (HER2-low group: median OS, 15.4 months vs. 11.5 months; hazard ratio, 0.75; HER2-zero group: median OS, 13.6 months vs. 10.8 months; hazard ratio, 0.85) [71]. SG thus demonstrated effectiveness regardless of the HER2-IHC status, both in HR+ breast cancer and TNBC. Building on these findings, the U.S. FDA and European Medicines Agency have approved the use of SG for treating pretreated metastatic TNBC or HR+/HER2-negative breast cancer.
3. Optimal treatment sequence of T-DXd and SG in the treatment of advanced HER2-low breast cancer
1) HR+/HER2-low breast cancer
Based on the inclusion criteria of the DB-04 trial, T-DXd is a treatment option for patients previously treated with endocrine treatment and/or at least one line of chemotherapy in both the ESMO living guidelines (2023 May) and the National Comprehensive Cancer Network (NCCN) guidelines (ver. 4.2023) [72]. Although there is no head-to-head comparison of SG and T-DXd in the current literature, T-DXd might be preferred in HR+/HER2-low breast cancer in a second-line setting because the patients who were included the DB-04 trial were a less pretreated population (1-2 prior lines of chemotherapy) than those in the TROPiCS-02 trial (2-4 prior lines of chemotherapy) [20,72]. If ADCs were not used in previous lines, T-DXd or SG are both considered available options in the third line and beyond because the optimal sequencing of ADCs has not been established yet.
2) HR–/HER2-low breast cancer
Although T-DXd was associated with a better PFS (hazard ratio, 0.46; 95% CI, 0.24 to 0.89) and OS (hazard ratio, 0.48; 95% CI, 0.24 to 0.85) among patients with HR–/HER2-low tumors than the physicians’ choice chemotherapy in the DB-04 trial, only 10% of the cohort in the DB-04 trial had HR– tumors (n=58). The phase III ASCENT trial included only patients with TNBC (n=468) and demonstrated the PFS and OS benefit of SG compared with the physician’s choice chemotherapy [67]. The primary outcome of the ASCENT trial was the efficacy of SG among patients with TNBC, and the evidence level for SG is considered higher than that for T-DXd in this population [20]. Nevertheless, it should be noted that there is no head-to-head comparison between T-DXd and SG in this population. Therefore, both agents could be considered as therapeutic options in second-line or subsequent settings in this population without germline BRCA1/2 mutations [72] and further studies will be needed to define the optimal sequence of these agents.
4. Intracranial activity of ADCs for HER2-low tumors
Brain metastasis (BM) remains a crucial concern in breast cancer; however, it has not been extensively studied in the context of HER2-low breast cancer to date. In addition to conventional treatments such as surgery, whole brain radiation, and stereotactic radiosurgery, there is a critical need for more effective systemic therapies for the optimal management of BM. Therefore, assessing the intracranial activity of ADCs for HER2-low tumors is of paramount importance.
The intracranial activities of T-DXd have been assessed in patients with HER2-positive breast cancer, as reported in trials such as the TUXEDO-1 trial [73], DEBRRAH trial [74], and DESTINY-Breast03 trial [75]. However, there is limited available data confined to HER2-low tumors. In the preclinical models, T-DXd reduced tumor growth and prolonged OS in a HER2-low BM patients–derived xenograft model [76]. In terms of clinical studies, T-DXd had an intracranial ORR of 33.3% in patients with HER2-low tumors and BMs progressing after prior treatment (cohort 4, where 2 out of 6 had an intracranial PR) in the phase II DEBRRAH trial [77]. Similarly, the best overall response of the intracranial lesions in HER2-low cohort in the DAISY trial was 30%, and the intracranial clinical benefit rate [CBR] was 50% [78]. Recently, a subgroup analysis of patients with BM from the DB-04 study was reported [79]. At the baseline, 35 patients had asymptomatic BM, and among these, 24 patients were randomized into the T-DXd arm while 11 patients were randomized into the physician’s choice chemotherapy arm. Among them, eight and seven patients in the T-DXd and control arms, respectively, had BMs, which were not previously treated. Four patients had an intracranial complete response [CR] and two patients had an intracranial PR in the T-DXd arm, whereas none of the patients in the control arm had a CR or PR. Similarly, the intracranial CBR and intracranial DCR were higher in the T-DXd arm (58.3% vs. 18.2% for CBR; 75.0% vs. 63.6% for DCR). Taken together, these findings suggested that T-DXd might have meaningful intracranial efficacy in HER2-low tumors. Due to the limited number of patients, further large-scale clinical trials will be warranted to evaluate the intracranial activity of T-DXd among patients with HER2-low tumors.
Regarding SG, there is limited evidence of intracranial activity among patients with HER2-low tumors. In the ASCENT trial, 61 patients with asymptomatic BM were included (32 in the SG arm, and 29 in the physician’s choice chemotherapy arm), with the intracranial ORR being 3% (1/32) in the SG arm and 0% in the control arm, as determined by a central review [80]. The results of ongoing clinical trials investigating the efficacy of SG among patients with BMs (SWOG-S2007 and NCT04647916) will clarify the intracranial efficacy of SG among patients with HER2-negative tumors (including HER2-low tumors) and BMs.
5. Evaluation of discordance of HER2 status for T-DXd treatment
As described above, HER2-low status seems to be unstable and changes dynamically during the course of the disease. In the DB-04 trial, HER2 status was defined using the primary tumor (35%) and metastatic tissue (65%) [2]. The efficacy of T-DXd was consistent regardless of the tumor sample characteristics or whether the HER2 status was determined with the primary tumor or metastatic tissue. Any point of results regarding HER2 status are acceptable for decision-making to use T-DXd. In addition, if the tumor tissues at different timepoints or lesions show conflicting results (e.g., one lesion is HER2-low and another lesion is HER2-zero), treatment with T-DXd might be possible [20]. Repeated biopsy and re-evaluation of HER2 IHC status at the timepoint of the development of metastasis might help to identify eligible candidates for T-DXd.
Treatment of HER2-Low Tumors in the (Neo) Adjuvant Setting
No anti-HER2-targeting agent is currently approved as neoadjuvant or adjuvant chemotherapy for early HER2-low breast cancer. There are several ongoing clinical trials to evaluate the efficacy of anti-HER2-targeting agents in early settings. The details of these clinical trials are summarized below (Table 3).
Ongoing Clinical Trials and Novel Agents
1. T-DXd
The DESTINY-Breast06 (DB-06, NCT04494425) study is a randomized phase III trial comparing the efficacy of T-DXd and physician’s choice of chemotherapy among patients with HR+/HER2-low, chemotherapy-naïve, endocrine-refractory advanced/mBC. Notably, a central review of HER2 IHC status is mandatory in the DB-06 trial, making the results regarding HER2 IHC status more reliable. In addition, the DB-06 trial allows the enrollment of patients with ‘HER2-ultralow,’ which is defined by IHC > 0 but < 1+ on central assessment. Hence, the results of the DB-06 trial are anticipated to offer valuable insights into the effectiveness of T-DXd because of the inclusion of not only patients with HER2-low but also those with extremely low HER2 expression.
The DESTINY-Breast15 trial (DB-15, NCT05950945) is a randomized, phase III trial to evaluate the efficacy of T-DXd among previously treated patients with unresectable, metastatic HER2-low or HER2-zero breast cancer. The patients are categorized into four cohorts according to their HR and HER2 IHC status. The primary endpoint of this study is the time to subsequent treatment to measure how long a clinical benefit is derived from T-DXd. The results of the DB-15 study will provide information on the clinical benefits of T-DXd in HER2-zero breast cancer.
Research on combining T-DXd with other anticancer drugs is ongoing. The DESTINY-Breast08 trial (DB-08, NCT0455-6773) is a phase Ib trial evaluating the efficacy and safety of a combination of T-DXd and other anticancer agents, including cytotoxic chemotherapeutic agents (capecitabine and paclitaxel), checkpoint inhibitors (durvalumab), a tyrosine kinase inhibitor (AKT inhibitor), and endocrine treatment (fulvestrant and anastrozole). Furthermore, a combination of T-DXd and valemetostat (a dual inhibitor of enhancer of zeste homolog 1 and 2 [EZH1/2]) is being studied among patients with HER2-low, HER2-ultralow, and HER2-zero mBC (NCT05633979).
In early-stage HER2-low breast cancer, a combination of T-DXd with immunotherapy (durvalumab) [NCT05795101, TRUDI study] or endocrine therapy (anastrozole) [NCT0455- 3770, TRIO-US B-12 TALENT trial] as neoadjuvant chemotherapy is under evaluation. In the phase II TALENT trial, patients with operable invasive HR+/HER2-low breast cancer were randomized 1:1 to neoadjuvant T-DXd monotherapy or T-DXd in combination with anastrozole. The interim results reported at the 2022 San Antonio Breast Cancer Symposium demonstrated that the addition of endocrine therapy to T-DXd did not enhance treatment efficacy. The overall response rate was 75% (12/16) in the monotherapy arm compared with 63.2% (12/19) in the combination arm. Only one patient in the monotherapy arm achieved a pCR, and none of the patients in the combination arm did [81]. Although the pCR rate was low, neoadjuvant T-DXd showed promising clinical activity in patients with HR+/HER2-low breast cancer. This is the first report of neoadjuvant T-DXd being used for patients who are currently being treated with neoadjuvant anthracycline/taxane-based chemotherapy. The updated final results are eagerly anticipated.
2. Sacituzumab govitecan
The ASCENT-07 trial (NCT05840211) is a phase III trial evaluating the efficacy of SG versus physician’s choice chemotherapy among patients with advanced/metastatic endocrine-refractory, HR+/HER2-negative breast cancer. In contrast to the TROPiCS-02 trial, which includes heavily treated patients (3rd line or later), the ASCENT-07 trial only includes chemotherapy-naïve patients in advanced/metastatic settings. The results of the ASCENT-07 trial and DB-06 trial are expected to provide information on the role of ADCs in patients with endocrine-refractory, chemotherapy-naïve HR+/HER2-low breast cancer.
3. Novel agents
1) Datopotamab deruxtecan (Dato-DXd)
Datopotamab deruxtecan (Dato-DXd) is a TROP2-directed ADC composed of a humanized anti-TROP2 IgG1 mAb (similar to SG) and deruxtecan (similar to T-DXd) [82]. In the phase III TROPION-Breast01 study, Dato-DXd demonstrated improved PFS compared to the investigator’s choice chemotherapy (median, 6.9 vs. 4.9 months; HR, 0.63; p < 0.0001) among patients with HR+/HER2-negative mBC (including HER2-low tumors) who had previously received 1-2 lines of chemotherapy [83]. Furthermore, the incidence of grade 3 or higher toxicity in the Dato-DXd group was lower than in the control arm, suggesting the safety of Dato-DXd. Similarly, in the TROPION-PanTumor01 study, Dato-DXd showed clinical efficacy among the heavily treated TNBC patients (n=44) with an ORR of 32% (1 CR, 13 PRs) and a DCR of 80% [84]. In addition, the combination of Dato-DXd and durvalumab (arm 7 in the BEGONIA study) had an ORR of 79% (6 CR, 43 PRs) and a median PFS of 13.8 months (95% CI, 11.0 to not calculated) in first-line advanced/mBC [85]. Building upon these results, the phase III TROPION-Breast02 study (NCT05374512) is presently underway, focusing on patients with previously untreated TNBC who are not eligible for immunotherapy [86].
In conclusion, Dato-DXd appears to be an effective and safe treatment option among patients with advanced HR+/HER2-negative breast cancer and TNBC. Specific analyses for HER2-low breast cancer are not currently available; thus, further subgroup analyses will be helpful for a better understanding of the efficacy of this drug for HER2-low breast cancer.
2) Trastuzumab-duocarmazine (SYD985)
Trastuzumab-duocarmazine (SYD985) is an ADC composed of trastuzumab, duocarmycin (payload), and a cleavable linker. Duocarmycin is believed to bind to the minor groove of DNA, causing irreversible DNA alkylation and ultimately leading to cell death [87]. In the phase III TULIP trial, trastuzumab–duocarmazine was associated with a better PFS (median, 7.0 months; HR, 0.64; p=0.002) than the physician’s choice chemotherapy in previously treated (≥ 2 lines) HER2-postive mBC [88]. The efficacy of trastuzumab–duocarmazine was also shown in patients with HER2-low breast cancer. In a phase I trial for patients with locally advanced or metastatic solid tumors, 49 patients with HER2-low tumors (32 patients with HR+ tumors and 17 with HR– tumors) were enrolled. The median number of prior chemotherapy lines was seven and four for patients with HR+ and HR– tumors, respectively. The ORR was 28% and 40% for patients with HR+/HER2-low and HR–/HER2-low tumors, respectively [89]. To date, none of the ongoing clinical trials with trastuzumab–duocarmazine have specifically focused on HER2- low breast cancer.
3) SHR-A1811
SHR-A1811 is an ADC including trastuzumab, a topoisomerase I inhibitor payload (SHR9265), and a cleavable linker with a DAR of 5.7 [90]. In in vivo and in vitro studies, SHR-A1811 showed high membrane permeability and cell-killing activity [90]. In addition, in a phase I clinical trial (NCT0444620) including different types of solid cancers, the incidence of grade 2 interstitial lung disease was less than 2.5%, and the treatment discontinuation rate was 5.1% across doses [90]. Based on these findings, SHR-A1811 is now being studied in phase II and III trials with or without other anticancer drugs for HER2-low breast cancer (NCT05018676, NCT05792410, NCT05845138, and NCT05814354) and other solid cancers including gastric cancer, colorectal cancer, and non–small cell lung cancer (NSCLC).
In the early setting, neoadjuvant SHR-A1811 is under investigation (NCT05911958). The primary outcome of this study is the ORR.
4) ARX788
ARX788 is an ADC composed of a humanized HER2-targeting mAb and a cytotoxic tubulin inhibitor (amberstatin, AS269), with a DAR of 1.9 [91]. ARX788 has shown antitumor efficacy in a preclinical model of HER2-positive and HER2-low breast cancer [91]. In a recent phase I trial that included patients with HER2-positive mBC, ARX788 exhibited clinical efficacy with an ORR of 65.5% (19/29) at the recommended dose (1.5 mg/kg) and a median PFS of 17.02 months (95% CI, 10.09 to not reached) [92]. Building upon these findings, ARX788 is currently under investigation in patients with previously treated (≥ 2 lines) HER2-low advanced breast cancer through a phase II, single-arm clinical trial design (NCT05018676).
5) Disitamab vedotin (RC-48, DV)
Disitamab vedotin (DV) is an ADC composed of hertuzumab (a novel anti-HER2 mAb), an anti-mitotic activity payload–monomethyl auristatin E (MMAE), and a cleavable linker. DV has shown antitumor efficacy for HER2-positive gastric cancer in a phase I trial (NCT02881138), with an ORR of 21% (12/57) [93]. Notably, patients with HER2-IHC2+without gene amplification responded similarly, with an ORR of 35.7% (5/14) [93]. DV also showed good efficacy at 2.0 mg/kg doses among patients with HER2-positive (n=70) and HER2-low (n=48) heavily treated mBC. In a HER2-positive cohort with a 2.0 mg/kg dose, the ORR was 42.9% (95% CI, 21.8 to 66.0) and the median PFS was 5.7 months (95% CI, 5.3 to 8.4). In the HER2-IHC2+ without gene amplification cohort, the ORR was 42.9% and the median PFS was 6.6 months (95% CI, 4.1 to 8.5). For patients with IHC1+ tumors, the ORR and median PFS were 30.8% (4/13) and 5.5 months (95% CI, 2.7 to 11.0), respectively. Most of the AEs were grade 1 or 2 [94]. Based on these findings, several phase II and III clinical trials using DV are ongoing among patients with advanced HER2-low breast cancer (NCT05904964, NCT05831878, and NCT04400695).
In the early stage, a combination of DV and penpulimab [95] (an anti–programmed death-1 [PD-1] mAb) is being studied in the neoadjuvant setting for HER2-low breast cancer (NCT05726175). Furthermore, neoadjuvant DV followed by tislelizumab (an anti–PD-1 mAb) is under evaluation (NCT05861635).
6) Zenocutuzumab (MCLA-128)
Zenocutuzumab is a bispecific humanized immunoglobulin G1 mAb against HER2 and HER3 [96]. Zenocutuzumab binds to the HER2 protein located on the cell surface (acting like a docking site) and interferes with NRG1 binding to HER3, inhibiting phosphorylation of HER3 and downstream oncogenic signaling [97]. Recently, zenocutuzumab has shown durable efficacy among NRG1 fusion-positive heavily treated solid cancer [98] and has been granted breakthrough therapy designation in NSCLC and pancreatic cancer by the U.S. FDA. In cases of HER2-low breast cancer, a combination of zenocutuzumab and endocrine therapy showed clinical activity as a DCR of 45% (95% CI, 32 to 59) and tolerable safety profiles among patients with HR+/HER2-low mBC who had progressed with a CDK4/6 inhibitor [99]. A phase II trial is currently ongoing to evaluate the efficacy of zenocutuzumab alone and in combination with other drugs (trastuzumab, vinorelbine, and endocrine therapy) among patients with HER2-positive and HR+/HER2-low mBC (NCT03321981).
7) Cancer vaccines
Nelipepimut-S (E75) is a peptide vaccine targeting HER2. In a phase II trial comparing trastuzumab and trastuzumab plus nelipepimut-S as adjuvant chemotherapy among HER2- low breast cancer, the combination group did not show any benefit of adding nelipepimut-S in the intention-to-treat analysis [100]. Recently, the authors reported the results of a subgroup analysis of patients with HR– subset (considered as TNBC). In this subgroup analysis (n=99), patients with HER2-IHC1+ (hazard ratio, 0.17; 95% CI, 0.04 to 0.79), those with HLA-A24 positivity (hazard ratio, 0.08; 95% CI, 0.01 to 0.61; p < 0.01), and those who received neoadjuvant chemotherapy (hazard ratio, 0.21; 95% CI, 0.06 to 0.076; p < 0.01) had improved 3-year DFS compared with the placebo group [101]. Although the number of participants was limited, these results might warrant further investigation of nelipepimut-S among TNBC subset.
AST-301 is a DNA-based cancer vaccine encoding the HER2 intracellular domain sequence. AST-301 has shown clinical efficacy and safety in a HER2-positive breast cancer population (NCT00436254) and HER2/neu-expressing gastric cancer xenograft model [102]. The Cornerstone-001 study is ongoing to evaluate whether adding AST-301 to standard adjuvant chemotherapy could reduce recurrence in HER2-low breast cancer (NCT05163223).
Other novel agents also currently being investigated for HER2-low breast cancer include MRG-002 (NCT04742153), pyrotinib (NCT05806671 and NCT05165225), and A166 (NCT03602079) (Table 3).
Future Directions
1. Applications of HER2-targeting ADCs for HER2-zero breast cancer
In addition to HER2-low tumors, previous data have suggested that some patients with HER2-zero breast cancer according to the current ASCO/CAP guidelines may benefit from HER2-targeting ADCs. In the phase II DAISY trial, the authors reported that some patients with HER2-zero (IHC0) showed a PR (ORR of 29.7%) after treatment with T-DXd. Notably, the authors conducted further analysis of whether the levels of HER2 expression (measured by reverse transcription polymerase chain reaction) could predict the response of HER2-zero breast cancer to T-DXd. In the results, the ORRs were 35.7% (5/14) and 30% (3/10) in patients with ERBB2 expression below and above the median, respectively. In addition, when HER2-stained slides in cohort 3 were rereviewed by two pathologists, 15 samples showed detectable HER2 expression (8 ‘ultralow’ and 7 IHC1+). The ORRs of samples with or without detectable HER2 expression were 40% (6/15) and 25% (4/16), respectively, in cohort 3. No significant differences in responses were observed according to ERBB2 gene expression in patients with HER2-IHC0, suggesting potential T-DXd efficacy in these patients regardless of ERBB2 expression. These findings warrant future ADC clinical trials among patients with HER2-zero breast cancer.
2. Limitations of current guidelines and new measurement methods for HER2 expression
Indeed, the current ASCO/CAP guidelines for HER2 status evaluation are intended to identify HER2-positive cases rather than to distinguish between HER2-low and HER2-zero tumors. Thus, there are some challenges in current HER2 testing for identifying HER2-low tumors, such as inter-laboratory variability and interpreter variability. The concordance rate of distinguishing IHC0 from IHC1+ was only 26% among the 18 experienced pathologists in Fernandez et al.’s report [103], and Schettini et al. [11] reported that 35% of cases in the HER2-low subgroup were discordant among five pathologists. Although IHC is easy to perform and cost-effective [104], the semi-quantitative nature of its interpretation can lead to differences in observations between pathologists [103,105].
Due to the low concordance rate among pathologists, new guidelines are needed to accurately detect HER2-low breast cancer since the introduction of T-DXd for HER2-low breast cancer. The ASCO and CAP recently updated their HER2 guidelines to address these concerns [106]. The updated guidelines suggest the use of the prior 2018 ASCO/CAP HER2 guideline recommendations [107], and no new reporting terminology, such as HER2-low or HER2-ultralow, has been adopted due to a lack of evidence to support whether HER2-low breast cancer is a distinct entity. However, the guidelines also recommend that pathologists should add a footnote regarding the specific IHC results to help identify eligible candidates for T-DXd among patients with HER2-negative breast cancer.
In addition, quantitative measurements of HER2 expression might be helpful to overcome this problem. Quantitative continuous scoring (QCS) is a deep learning-based image analysis model that can automatically determine HER2 expression. In a validation analysis, QCS demonstrated a high correlation with pathologist results (R=0.993) [108], mRNA quantification of expression of the ERBB2 gene (R=0.81), and the immunofluorescence H-score (R=0.85) [109]. Notably, QCS could detect the spatial distribution of HER2 expression in HER2-negative breast cancer [110] and stratified the HER2-low population treated with T-DXd (n=64) into two subgroups with distinct ORR and PFS rates based on high and low QCS [108]. These results indicate that HER2 QCS might be useful for identifying potential candidates for T-DXd in the HER2-low group and predicting their prognosis.
Recently, Moutafi et al. presented a new quantitative method for detecting HER2-low tumors that involves a combination of a quantitative immunofluorescence- and mass spectrometry-standardized HER2 array on conventional histology slides [111]. This new method, the HER2-V2 assay, can detect HER2 expression below the threshold of conventional HER2 testing and quantitatively present the results. When applying the HER2-V2 assay to 364 breast cancer cases, 67% of patients exhibited HER2 expression that was above the quantification limit but below the levels typical of HER2-amplified breast cancer [111]. Although this method has several limitations such as strong dependence on tissue microarrays, it might be helpful for a more objective assessment of low levels of HER2 expression and the selection of candidates for anti-HER2 ADCs, and thus warrants further investigation.
3. Challenges in selecting and sequencing ADCs for optimal outcomes
Over the past 20 years, ADCs have undergone significant evolution, encompassing target diversity, payload ratio, and linker design, leading to a progressive increase in their efficacy. There are still unanswered clinical questions regarding the optimal sequencing of treatment for patients eligible for more than one ADC and, in general, how to manage patients beyond progression on ADCs. Resistance mechanisms to ADCs may include target resistance and payload resistance [112]. The emergence of novel ADCs has presented challenges, particularly related to certain side effects, such as interstitial lung disease with T-DXd, necessitating awareness and efforts for mitigation [113]. As the AEs of ADCs vary based on antibodies and payloads, the individual patient characteristics and unique AEs of each drug should be considered when selecting optimal ADCs. With the shift of ADCs into earlier lines of therapy, possibly even into the early-stage setting, learning to use ADCs in sequence becomes crucial. This will require high-quality real-world evidence, prospective studies, and large translational analyses to determine the key biomarkers and factors for predicting responses, treatmentrelated AEs, and mechanisms of resistance to both the mAbs and their payloads.
Conclusion
Based on the current evidence, HER2-low breast cancer is not considered a distinct breast cancer subtype. However, recent clinical trials have indicated that HER2-low breast cancer holds potential as a targetable subset for anti-HER2 ADCs, including T-DXd, SG, and other novel HER2-targeting ADCs. The findings from ongoing clinical trials are anticipated to offer valuable insights into refining the optimal treatment strategies for HER2-low breast cancer. In addition, further investigation will be necessary to assess the effectiveness of HER2-targeting ADCs in patients with HER2-ultralow and HER2-zero (IHC0) tumors.
XML Download