Breast cancer (BC) was the most frequent tumor and the leading cause of cancer-related deaths among women, with more than 2,3 million cases worldwide in 2022 (1). BC is highly heterogeneous at morphologic and molecular levels, which directly impacts the disease prognosis and treatment (2). Four main intrinsic BC subtypes have been well-characterized and comprise the vast majority of the biological diversity in BC: Luminal A, Luminal B, human epidermal growth factor 2 (HER2) enriched (HER2+), and Basal-like (3). These subtypes differ regarding the proliferation/cell cycle-related and luminal/hormone-regulated pathways, gene expression regulation mechanisms, such as methylation, and mutational burden (4).

HER2, also known as Neu or ErbB2, belongs to the ERBB receptor family, which is expressed in many epithelial, mesenchymal, and neuronal cells, and has different roles in cellular development, proliferation, and differentiation (5). The HER2 protein composes a heterodimer with other ERBB and epidermal growth factor receptors (EGFRs). These complexes recognize ligand hormones and trigger subsequent signal transduction, activating downstream signaling pathways such as PI3K-AKT and MEK-ERK (6). HER2 overexpression is a negative prognostic factor and accounts for about 15–20% of all BC cases. One of the main mechanisms of HER2 activation is the amplification of the encoding gene ERBB2, leading to HER2 overexpression (7). This overexpression can be identified through the immunohistochemistry status (IHC3+), or in situ hybridization (ISH) measurement of an ERBB2 gene copy number of six or more, or through the ERBB2/CEP17 (centromeric region of chromosome 17) ratio of 2.0 or greater (8).

Most BC cases (approximately 80%) are not metastatic at the time of diagnosis. For these patients, the treatment strategy is based on tumor eradication and recurrence prevention (9). Hence, choosing effective drugs is essential for treatment success. Specific therapeutic strategies (both neoadjuvant and adjuvant therapies) were established based on BC molecular subtypes. For example, monoclonal antibodies (trastuzumab and pertuzumab) associated with conventional chemotherapeutic drugs are effective in the treatment of HER2+ BC patients (10). Although cancer therapeutic schemes are based on cancer staging and molecular subtypes, there is a broad spectrum of treatment responses, which may be influenced by other factors, such as genetic and epigenetic alterations.

In the last decades, several studies reported the landscape of germline variants in cancer predisposition genes as well as their impact on cancer risk. As an example, Li-Fraumeni syndrome (LFS), an autosomal dominantly inherited condition, is a rare hereditary cancer syndrome characterized by a high and early-onset cancer risk caused by pathogenic variants (PVs) in the tumor suppressor gene TP53 (11). Other well-known genetic alterations include PVs or likely pathogenic variants (LPVs) in the BRCA1 and BRCA2 genes, linked to hereditary breast and ovarian cancer syndrome (12). Moreover, these germline variants may be associated with a poorer BC prognosis (13, 14).

Even though genomic studies have characterized the germline architecture of BC patients (15, 16), little is known regarding the impact of these variants on the predisposition and prognosis of specific BC subtypes. Thus, this study aimed to perform a comprehensive assessment of rare germline variants associated with HER2+ BC. To this end, we conducted a systematic review and meta-analysis of genomic, exomic, and panel sequencing studies that assessed germline variants associated with the prediction of HER2+ BC, in comparison to other BC subtypes.

It is known that approximately 10% of BC cases result from hereditary causes (33). Genetic testing has been widely implemented in BC care to determine hereditary cancer syndromes and personalized medicine (34). Here, we investigated the presence of rare germline variants in a specific group of genes with the aim of determining their possible association with HER2+ BC. We conducted a thorough systematic review and meta-analysis and compared the prevalence of these variants in subtypes characterized by HER2 overexpression with other well-defined BC subtypes. Of the 36 studies included in our analysis, 11 provided data on the distribution of germline variants in different clinically relevant BC subtypes, allowing comparative assessments. Our meta-analysis revealed significant differences in the occurrence of germline variants between BC groups for eight genes: BRCA1, BRCA2, TP53, ATM, CHEK2, PALB2, RAD51C, and BARD1.

In our analysis, we identified genes with germline variants that predispose to HER2 overexpressing subtypes: TP53, ATM, and CHEK2. Of these genes, germline variants on TP53 gene were found to be exclusive when comparing HER2+ and HER2- subtypes, supporting a link between germline variants in this gene and increased expression of the HER2 protein. In fact, the proportion of HER2+ BC is increased in patients with LFS (35–38). Breast tumors in TP53 germline variant carriers are usually high-grade, HER2+, and HR+ with a dense sclerotic tumor stroma (37).

The ATM germline variants were also significantly associated with subtypes overexpressing HER2 and/or HR, which is consistent with previous studies describing intermediate- to high-grade, HR+ disease and potentially higher rates of HER2 positivity and lymph node involvement associated with ATM germline variants (39, 40), as well as reduced recurrence time in invasive HER2+ BC patients (39). ATM truncating and missense variants are linked to an increased risk of estrogen receptor-positive (ER+) BC rather than ER- disease (41).

The CHEK2 germline variants were found associated with both HR+ and HER2+ status. The highest chance was found when comparing the HR+HER2+ subtype with TNBC. A study investigating CHEK2 germline variants identified an association of the immunophenotypic molecular subtypes of breast cancer with the type of the genetic variant. CHEK2-truncating variants were found to increase 6-fold the risk of luminal B (ER+ and/or PR+, HER2+) subtype (42). Moreover, other studies investigating the relation between CHEK2 germline pathogenic variants and receptor status do not show concordance. Variants in the CHEK2 gene were found to be associated with negative or positive status of HR and HER2 (43–45).

Remarkably, the TP53, ATM, and CHK2 proteins work together to control the cell cycle. When DNA damage occurs, ATM phosphorylates CHK2, triggering its activation. This in turn initiates a cascade of events that leads to the phosphorylation of various downstream substrates, including p53, which arrests the cell cycle and regulates apoptosis, and BRCA1, which modulates DNA repair (46). Interestingly, HER2 blockade indirectly affects downstream signaling pathways such as AKT, which could have secondary effects on CHK2 activity. In this sense, an effective treatment strategy targeting HER2 blockade would lead to inhibition of the AKT signaling pathway, thereby abrogating its inhibition on CHK2. This disruption would facilitate CHK2 function and enable p53-mediated repair of DNA damage. However, mutations in CHK2, p53, or ATM could impair this response by preventing CHK2 activity directly or by inhibiting CHK2 phosphorylation by ATM (47).

The present study confirms that BRCA1 germline variants are consistently associated with a higher susceptibility to TNBC, with a lifetime risk of between 50 and 85% (48–50). In addition, BRCA2 germline variants show an increased predisposition for HER2- groups (48, 49). Furthermore, PALB2 germline variant carriers were found to have a higher predisposition to develop TNBC. In fact, BC in PALB2 germline variant carriers has similar phenotypic characteristics to our results: 40% are TNBC, and 93% are HER2- (51). Moreover, germline variants on BARD1 and RAD51C genes were found to predispose to TNBC. Interestingly, the BARD1 protein interacts with BRCA1 (BRCA1-BARD1) forming a tumor suppressor complex, which is an E3 ubiquitin ligase necessary for DNA double-strand breaks repair by RAD51-mediated homologous recombination (52). Identifying the three genes predisposing to TNBC may suggest a role of this pathway in the BC subtype development.

This study has some limitations. Smaller studies reporting lower OR for rare germline variants in individuals with BC are underrepresented in the scientific literature, possibly indicating a potential bias toward publishing studies with stronger or more significant findings. To address this bias, this work included only studies that perform multi-genic analysis. In addition, some publications focused only on a limited number of variants within their cohorts. Despite our efforts to ask authors for additional data for subgroup analyses, the meta-analyzed data were ultimately limited compared to larger cohorts. A risk of bias assessment revealed that confounding factors were identified in several studies, with no methods provided to address them. However, given the nature of the data we used for the meta-analysis, these confounders would not directly interfere with our analyses.

Finally, this study strengthens our understanding of the gene-specific risks associated with HER2 overexpression and provides crucial insights to identify genes that need to be tested in the context of HER2+ BC. Such findings could improve the management of hereditary BC by guiding the exploration of genes and prognosis of the disease through the correlation of gene-subtype BC. To our knowledge, this is the first meta-analysis to investigate genetic variants associated with HER2+ BC, through comparisons between BC subtypes. The meta-analysis results and literature review findings point to the importance of a closer examination of TP53 germline variants in relation to breast tumors with overexpressed HER2. Also, CHEK2 and ATM germline variants predisposed to HER2+ only when compared to TNBC and could be involved with HER2 overexpression more intricately. Furthermore, variants in BRCA1 may confer a higher risk for the development of TNBC in comparison to other subtypes. This association was already suggested in the literature and the meta-analysis was able to confirm the evidence. Moreover, BRCA2 germline variants exhibited a greater predisposition to HER2- when compared to HER2 overexpressing groups, implying a possible involvement of this gene in the HER2 expression.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

AB: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – original draft. NC: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – original draft. LP: Data curation, Formal analysis, Investigation, Methodology, Writing – original draft. GC: Data curation, Formal analysis, Investigation, Writing – review & editing. JS: Investigation, Methodology, Writing – review & editing. MB: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Writing – review & editing. CB: Funding acquisition, Validation, Writing – review & editing. MR: Funding acquisition, Validation, Writing – review & editing. GM: Conceptualization, Funding acquisition, Methodology, Supervision, Writing – review & editing. DR: Conceptualization, Funding acquisition, Methodology, Supervision, Writing – review & editing.