Abstract
Fibroblast growth factor receptor 1 (FGFR1) amplification, a significant cancer alteration with both prognostic and therapeutic relevance, lacks a standardized definition. The relationship between FGFR1 copy number (CN) variations and tumour microenvironment (TME) characteristics also remains poorly understood. This study defined FGFR1 amplification and investigated TME features across FGFR1 CN subtypes in triple‐negative breast cancer (TNBC) to guide precise treatment. We analysed 182 TNBC tumour specimens. FGFR1 CN status was assessed using fluorescence in situ hybridization (FISH). Next‐generation sequencing (NGS) was performed on 88 samples to explore biological differences. Immunohistochemical (IHC) results for CD3, CD4, and CD8 along with clinical data were analysed. Samples were categorized into four groups based on the FGFR1/CEN8 ratio and the mean FGFR1 signals per cell. Groups 1 (ratio ≥ 2.0, signals ≥ 4.0) and 2 (ratio ≥ 2.0, signals < 4.0) exhibited robust tumour proliferation, increased FGFR signalling, and an immunosuppressive TME. Conversely, group 4 (ratio < 2.0, signals < 4.0) showed significantly enhanced immune cell infiltration (e.g. CD8+ T cells, NK cells), supported by elevated CD3+/CD8+ T‐cell densities. Group 3 (ratio < 2.0, signals ≥ 4.0) was heterogeneous: tumours with FGFR1 CN ≥ 5 resembled groups 1/2, while those with CN < 5 mirrored group 4. Thus, FGFR1 amplification (FGFR1‐Amp) was defined as FGFR1/CEN8 ratio ≥ 2.0 or CN ≥ 5.0 (14.29% prevalence, 26/182). Remaining cases were FGFR1‐neutral (FGFR1‐Neu). Immune profiling revealed that FGFR1‐Amp & HER2‐low TNBC had reduced immune cell infiltration but heightened oncogenic signalling (including epithelial–mesenchymal transition, angiogenesis) compared with other subgroups. In contrast, FGFR1‐Neu tumours, especially HER2‐zero, displayed increased immune cell infiltration (NK cells, effector cells), MHC I/II molecules, activation markers, and elevated immune checkpoint expression (e.g. PD‐L1, CTLA4, LAG3). This study defines FGFR1 amplification in TNBC and characterizes its TME. FGFR1‐Amp in HER2‐low TNBC correlates with an immunosuppressive environment, potentially hindering HER2‐ADC efficacy and suggesting combined FGFR inhibition. Conversely, FGFR1‐Neu & HER2‐zero tumours with immune activation may respond better to immunotherapy. © 2026 The Author(s). The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.
Keywords: triple‐negative breast cancer (TNBC), fibroblast growth factor receptor 1 (FGFR1), HER2, amplification, tumour microenvironment, immune cell
Introduction
The fibroblast growth factor receptor 1 (FGFR1) gene encodes a receptor tyrosine kinase that initiates downstream signalling pathways, including MAPK/ERK and PI3K/AKT, which are integral to cellular proliferation, survival, and epithelial–mesenchymal transition (EMT) processes [1, 2]. The copy number (CN) alteration of FGFR1, which promotes tumour growth and reshapes the immune landscape, has emerged as a critical oncogenic event and a potential target in triple‐negative breast cancer (TNBC) [2, 3]. FGFR1 CN gain was previously observed in approximately 7–10% of breast cancer cases, with a notable prevalence in certain TNBC subtypes [1, 4]. Clinically, elevated FGFR1 CN is associated with poor overall survival in TNBC patients [5, 6]. Preclinical studies have demonstrated that FGFR1 signalling exacerbates the immunosuppressive tumour microenvironment through multiple mechanisms, which may represent a critical factor contributing to poor prognosis. In particular, FGFR1 activation in cancer‐associated fibroblasts enhances their proliferation and promotes the secretion of vascular cell adhesion molecule 1 (VCAM‐1), thereby establishing physical and chemical barriers that hinder T‐cell infiltration [3]. Furthermore, FGFR1 signalling upregulates the expression of immunosuppressive cytokines, such as interleukin‐10 (IL‐10) and transforming growth factor‐beta (TGF‐β), along with angiogenic factors such as vascular endothelial growth factor (VEGF), thereby promoting an environment that facilitates immune evasion [7, 8]. Despite significant advancements in understanding FGFR1 amplification in TNBC, the relationship between FGFR1 amplification and the TNBC immune microenvironment remains poorly characterized.
In spite of ongoing research, a universally accepted standard for defining FGFR1 amplification remains elusive. The most common definition of FGFR1 amplification requires either an FGFR1/centromere 8 (CEN8) ratio exceeding 2.0 or six or more FGFR1 copies per tumour nucleus [9, 10, 11]. However, other studies use different criteria, such as more than five FGFR1 signals per nucleus in > 50% of cancer cells or the presence of large gene copy clusters [12]. These inconsistent thresholds for copy number or ratios across studies result in varying reports of its prevalence and clinical significance [2]. Consequently, establishing consensus guidelines for FGFR1 amplification testing remains an imperative unmet need in precision oncology. Therefore, by integrating genomic and transcriptomic analyses, this study seeks to develop a standardized framework for defining FGFR1 amplification and to investigate the characteristics of the tumour microenvironment (TME) associated with different FGFR1 CN alterations. This methodology aims to improve diagnostic precision and guide therapeutic strategies for cancers involving FGFR1 modifications.
Materials and methods
Ethics approval and patient consent
This study was approved by the Institutional Review Board of Peking Union Medical College Hospital (approval number: ZS‐2834). Informed consent forms were obtained from all patients.
Patient cohort and sample classification
The study encompassed 182 TNBC samples with sufficient archived tissue available for FGFR1 fluorescence in situ hybridization (FISH) testing, collected between 2016 and 2018. Tissue microarrays (TMAs) were constructed from formalin‐fixed, paraffin‐embedded (FFPE) specimens and subsequently employed for FGFR1 FISH analysis. All haematoxylin and eosin (H&E)‐stained slides from each case were reviewed to map intratumoural heterogeneity, including architectural patterns and geographical locations. For each case, two tissue cores, each measuring 1 mm, were extracted from representative tumour regions (Figure 1A). Clinicopathological data were systematically collected. Following the 2023 American Society of Clinical Oncology/College of American Pathologists (ASCO/CAP) guideline update [13], HER2‐low was defined as IHC 1+ or 2+/ISH‐negative, while HER2‐zero was defined as IHC 0.
Figure 1.

Comprehensive genomic and clinical landscape of FGFR1‐altered TNBC. (A) The workflow of this study. (B) Representative FISH images of FGFR1 (red) and CEN8 (green). Samples were categorized into four distinct subtypes based on FGFR1 fluorescence in situ hybridization analysis. Images were taken at 1,000× total magnification. (C) Genetic alterations and clinical characteristics in patient cohorts. (D) Correlation between intratumoural heterogeneity (ITH), mutant subclonal allele frequency (MSAF), tumour mutation burden (TMB), and FGFR1 status. (E) Association between FGFR1 status and mutation signatures. p values of statistical significance are shown in bold. TNBC, triple‐negative breast cancer; FISH, fluorescence in situ hybridization; CEN8, centromere of chromosome 8. *Stratified by FGFR1 status alone (uses all 88 samples; n = 88). #Stratified by both FGFR1 and HER2 status (only primary tumours without neoadjuvant therapy were included, resulting in n = 78).
FGFR1 fluorescence in situ hybridization
The FGFR1 FISH assays were conducted on 4‐μm‐thick FFPE tissue sections using an FGFR1/CEN 8 Dual Color FISH Probe (ZytoLight; ZytoVision GmbH, Bremerhaven, Germany) following the manufacturer's protocol. The slides were scanned and analysed using an Olympus BX51 Fluorescence Microscope (Olympus, Shinjuku, Tokyo, Japan). Pictures were taken at 1,000× total magnification. For each tissue core, FGFR1 (red) and CEN8 (green) signals were enumerated in at least 20 tumour cell nuclei by two independent observers. Given the lack of a universally accepted standard for defining FGFR1 amplification, we adopted clinically validated HER2‐FISH criteria (ASCO/CAP guidelines) [13] as a reference framework for cohort stratification, combined with evidence from previous reports [12, 14]. We identified an FGFR1/CEN8 ratio of 2.0 and FGFR1 signals per cell of 4.0 and 5.0 as key cutoff points. For cases with values approaching these cutoffs, a re‐evaluation was performed, with at least 20 additional tumour cell nuclei recounted to ensure accuracy.
Immunohistochemical staining
Immunohistochemical staining was performed on 4‐μm paraffin sections using a Leica BOND‐III automated immunostainer (Leica, Bannockburn, IL, USA) or Dako Autostainer Link 48 (Agilent Technologies, Santa Clara, CA, USA). Primary antibodies were applied in accordance with the manufacturers' protocol, including CD3, CD4, CD8, AR, FOXC1, and DCLK1 (supplementary material, Table S1). Densities of CD3+, CD4+, and CD8+ tumour‐infiltrating lymphocytes per mm2 were calculated using Fiji (version 1.53c), an open‐source distribution of ImageJ, available at https://imagej.net/software/fiji/ (accessed 20 August 2024). TNBC subtyping was determined by immunohistochemistry for AR, CD8, FOXC1, and DCLK1. A cutoff of ≥ 10% positive tumour cells was used to define positivity for AR, FOXC1, and DCLK1, while CD8 positivity was defined as ≥ 20% positive cells among all nucleated cells. Based on these markers, TNBCs were classified as luminal androgen receptor (LAR: AR+), immunomodulatory (IM: AR−/CD8+), basal‐like immune suppressed (BLIS: AR−/CD8−/FOXC1+), mesenchymal‐like (MES: AR−/CD8−/FOXC1−/DCLK1+), or unclassifiable (UC: negative for all four markers) [15, 16].
Next‐generation sequencing
Nucleic acid was extracted from six to eight tissue sections (4–5 mm thick) of FFPE samples, depending on the tumour content for each of the cases. DNA was extracted using a QIAamp DNA FFPE Tissue Kit (QIAGEN, Hilden, Germany), while RNA was extracted using an RNeasy FFPE Kit (QIAGEN), following dewaxing of the tissue sections. Nucleic acid was quantified using a NanoDrop One Spectrophotometer (BioTek Instruments, Winooski, VT, USA), and the quality was assessed using a Quantus Fluorometer (Promega, Madison, WI, USA). DNA was fragmented into 200–250 bp fragments and then underwent procedures including indexing, end‐repairing, A‐tailing, adaptor ligation, and amplification using the NEBNext Ultra II DNA Library Prep Kit (New England BioLabs, Ipswich, MA, USA). RNA was fragmented, followed by reverse transcription, cDNA synthesis, and strand‐specific library preparation using the NEBNext Ultra II Directional RNA Library Prep Kit for Illumina. DNA and RNA libraries were captured separately using the AmoyDx Master Panel containing 571 genes for DNA‐based analyses (single nucleotide variant, indel, fusion, copy number variation, microsatellite instability, and tumour mutation burden) and 2,660 genes for RNA expression and fusion detection (supplementary material, Tables S2–S5). Library sequencing was performed on an Illumina NovaSeq 6000 sequencer (Illumina, Inc., San Diego, CA, USA), generating 2 × 150 bp paired‐end reads.
Genomic variation analysis
Intratumour heterogeneity (ITH), maximum somatic allele frequency (MSAF), and tumour mutation burden (TMB) were compared across different groups. To delineate the mutational signatures of gene mutations, we used the SigMA (v 1.6) R package, which is a likelihood‐based measure combined with machine‐learning techniques [17].
Differential gene expression and pathway analysis
The Wilcoxon rank‐sum test, a classical non‐parametric statistical test, was used to compare gene expression levels between two conditions. Differentially expressed genes (DEGs) were identified based on the following criteria: genes with fold‐change ≥ 1.5 and p value < 0.05 for the comparison between FGFR1 groups. Enrichment analyses were conducted using the R package ‘clusterProfiler’ to determine the functional roles of DEGs. These included Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), Hallmark, WikiPathways, and Reactome. Gene set enrichment analysis (GSEA) was also performed by the R package ‘GSEA’.
Tumour microenvironment analysis
The gene set variation analysis (GSVA) algorithm was used to evaluate the relative abundance of infiltrating immune cells in the TME. Gene sets for TME‐infiltrating immune cells were extracted from previous datasets [18]. Enrichment scores calculated using the R package ‘GSVA’ were used to determine the relative abundance of each TME‐infiltrating cell [19]. The deconvolution method CIBERSORT [20] and marker‐based approach MCP‐counter [21] were also used to assess the TME.
Statistical analyses
Statistical analyses were performed using R software version 4.1.2 (https://www.r-project.org/, accessed 10 February 2022), GraphPad Prism version 9.0.0 (GraphPad Software, LLC, Boston, MA, USA), and SPSS version 27.0.1 (IBM Corp., Armonk, NY, USA). Fisher's exact test was utilized to compare unordered categorical variables. For continuous variables, we used the Mann–Whitney Wilcoxon test, while the Kruskal–Wallis test was employed for discrete categorical variables. A p value less than 0.05 was considered statistically significant.
Results
Distribution and clinicogenomic characteristics of FGFR1 copy number subgroups in TNBC
One hundred and eighty‐two samples (Table 1) were categorized into four groups based on the FISH results: (1) group 1 (FGFR1/CEN8 ratio ≥ 2.0 and mean FGFR1 signals per cell ≥ 4.0), (2) group 2 (FGFR1/CEN8 ratio ≥ 2.0 and mean FGFR1 signals per cell < 4.0), (3) group 3 (FGFR1/CEN8 ratio < 2.0 and mean FGFR1 signals per cell ≥ 4.0), and (4) group 4 (FGFR1/CEN8 ratio < 2.0 and mean FGFR1 signals per cell < 4.0). The distribution of the samples was as follows: group 1 comprised 8.2% (15/182) of the cohort; group 2 comprised 2.7% (5/182); group 3 comprised 17.6% (32/182); and group 4 comprised 71.4% (130/182) (Figure 1A,B). No significant differences were observed in the clinicopathological characteristics of TNBC patients with different copy number subgroups, except for Ki‐67 indices. Ki‐67 indices were significantly higher in group 1 and group 3 tumours compared with group 4 tumours (p = 0.047 and 0.038, respectively), while these differences were no longer statistically significant after adjustment (Bonferroni correction, adjusted p = 0.281 and 0.226, respectively) (supplementary material, Table S6). Notably, no statistically significant associations were observed between the FGFR1 copy number‐based subgroups and the conventional molecular subtypes of TNBC (Table 1; supplementary material, Figure S1). We next analysed genomic features across the four FGFR1 copy number subgroups in 83 tumours. Oncoplot analysis unveiled single‐nucleotide and copy number variation patterns (Figure 1C). The most prominent cancer‐related variations observed in this cohort were TP53 mutations (found in 73.5% of tumours), followed by KMT2C (28.9%), MUC16 (21.7%), and PIK3CA (19.3%) mutations. The top three somatic copy number variations (including amplification and deletion) with the highest frequency in patients affected the genes TOP2A (56.6%), CD274 (42.2%), and AURKA (33.7%), respectively. Notably, FGFR1 exhibited a 14.5% frequency of copy number alterations (amplification, 9.6%), and a positive correlation was observed between FGFR1 copy numbers determined by FISH and NGS (supplementary material, Figure S2). While no differences in ITH and MSAF were observed among the subgroups, TMB exhibited an upward trend in group 3 (p = 0.02) and group 4 (p = 0.048) when compared with group 1 (Figure 1D). Subsequent evaluation of genomic signatures demonstrated no statistically significant distribution differences among the subgroups (Figure 1E).
Table 1.
Baseline clinicopathological characteristics of triple‐negative breast cancer samples analysed by FGFR1 FISH and Master Panel sequencing.
| Variables, n (%) | FGFR1 FISH (n = 182) | Master Panel sequencing (n = 88) |
|---|---|---|
| FGFR1 | ||
| Group 1 | 15 (8.2%) | 15 (17.0%) |
| Group 2 | 5 (2.7%) | 5 (5.7%) |
| Group 3 | 32 (17.6%) | 28 (31.8%) |
| Group 4 | 130 (71.4%) | 40 (45.5%) |
| HER2 | ||
| Low | 84 (46.2%) | 35 (39.8%) |
| Zero | 98 (53.8%) | 53 (60.2%) |
| Histology | ||
| IBC‐NST | 172 (94.5%) | 85 (96.6%) |
| Other | 10 (5.5%) | 3 (3.4%) |
| Histological grade | ||
| Grade 1 | 4 (2.2%) | 1 (1.1%) |
| Grade 2 | 39 (21.4%) | 15 (17.0%) |
| Grade 3 | 133 (73.1%) | 70 (79.5%) |
| Unknown | 6 (3.3%) | 2 (2.3%) |
| Ki‐67 index (%, mean ± SD) | 55 ± 27 | 64 ± 24 |
| Tumour location | ||
| Primary tumour | 173 (95.1%) | 82 (93.2%) |
| Recurrent tumour | 7 (3.8%) | 4 (4.5%) |
| Lymphatic metastasis | 2 (1.1%) | 2 (2.3%) |
| Neoadjuvant therapy | ||
| Treated | 9 (4.9%) | 5 (5.7%) |
| Untreated | 173 (95.1%) | 83 (94.3%) |
| TNBC subtype (n = 166 or 79)* | ||
| IHC‐LAR | 55 (33.1%) | 18 (22.8%) |
| IHC‐IM | 43 (25.9%) | 24 (30.4%) |
| IHC‐BLIS | 47 (25.8%) | 26 (32.9%) |
| IHC‐MES | 3 (1.8%) | 3 (3.8%) |
| IHC‐UC | 18 (10.8%) | 8 (10.1%) |
BLIS, basal‐like immune‐suppressed; IBC‐NST, invasive breast carcinoma of no special type; IHC, immunohistochemistry; IM, immunomodulatory; LAR, luminal androgen receptor; MES, mesenchymal; UC, unclassifiable.
Some samples lacked sufficient tissue for IHC‐based molecular subtyping.
Differential enrichment of biological process among FGFR1 copy number subgroups
Eighty‐eight samples met quality criteria for comprehensive genomic profiling and targeted RNA sequencing (Table 1). Differential expression gene (DEG) analysis across the four FGFR1 copy number subgroups revealed divergent trends in tumour cell proliferation and invasion and immune‐activating molecules. The expression of genes related to tumour cell proliferation and invasion was found to be upregulated, such as AREG and GAB2 in group 1, and CEACAM6 and PI3K3CA in group 2. In addition, group 4 displayed significant upregulation of HLA‐DQB2, IGLL1, and CXXC4. In contrast, group 1 showed downregulation of CXCL13 and IFI27; group 2 exhibited reduced expression of HLA‐DRB4, HLA‐DQA2, and IGLL1. Group 3 demonstrated downregulation of HLA‐G, HLA‐DQB1, and HLA‐DQB2 (Figure 2A). Functional annotation of these DEGs was performed using GO, KEGG, Hallmark, WikiPathways, and Reactome analyses. Group 4 was predominantly enriched in immune activation pathways, such as leukocyte‐mediated immunity and T‐cell activation. Conversely, epithelial cell proliferation regulation, Notch signalling, and DNA repair pathways were enriched in group 1. Group 2 displayed pronounced enrichment of the PI3K–Akt, HIF‐1, and JAK–STAT signalling pathways, while group 3 showed particularly active cytokine networks, including cytokine–cytokine receptor interactions and the Ras signalling pathway (Figure 2B).
Figure 2.

Tumour microenvironment differences in FGFR1 copy number subgroups. (A) Differential gene expression and (B) pathway enrichment analysis across FGFR1 status groups. (C) The analysis utilizes the ESTIMATE algorithm to quantify the immune and stromal scores, tumour purity, and overall ESTIMATE scores for each FGFR1 status group. (D) Immune landscape and GSVA analysis across different FGFR1 status groups. (E) Expression levels of immune checkpoint molecules across different FGFR1 status groups. p values of statistical significance are shown in bold.
TME differences among FGFR1 copy number subgroups
We analysed RNA sequencing results to explore the immune microenvironment characteristics among the four FGFR1 copy number subgroups. The ESTIMATE analysis revealed a trend of increased stromal and immune scores, alongside decreased tumour purity in samples of group 4 compared with the other three groups (Figure 2C). Group 4 demonstrated significant infiltration of activated immune components, including cytotoxic CD8 T cells, NK cells, and effector populations, along with elevated immune activation signatures. Conversely, groups 1 and 2 had reduced immune cell recruitment and a suppressive microenvironment. Group 3 had a mixed immune profile, intermediate between the active state of group 4 and the depleted state of groups 1 and 2 (Figure 2D). Notably, groups 1 and 2 exhibited significant downregulation of activated immune populations (including activated CD8 T cells, M1 macrophages, and NK cells), whereas group 3 demonstrated selective enrichment of NK cells without comparable increases in other effector subsets (Figure 2D). Immune checkpoint gene expression analysis revealed that group 4 demonstrated elevated levels of CD274, HAVCR2, and PDCD1LG2, compared with the other three groups (Figure 2E). In terms of immunohistochemistry (IHC) staining intensity and cell density values, there was a tendency for more CD3+/CD4+/CD8+ T cells to infiltrate in group 4 (Figure 3A,B).
Figure 3.

Immunohistochemical validation analysis in FGFR1 copy number subgroups. (A) Representative IHC staining for CD3+/CD4+/CD8+ cells in the tumour immune microenvironment of each FGFR1 status group. To clearly illustrate the heterogeneity among the samples in group 3 (with different FGFR1 gene copy number states, CN ≥ 5 versus CN < 5), we present two representative cases separately. Scale bars: 200 μm. (B) Quantification of IHC positive CD3+/CD4+/CD8 cells across the FGFR1 groups.
FGFR1 copy number in FGFR1 group 3 impacts TME
At the genomic level, group 3 exhibits similarities to group 4 and is characterized by an elevated tumour mutational burden. Conversely, within the context of the tumour microenvironment (TME), the associated immune activation features in group 3 are neither downregulated, as observed in groups 1 and 2, nor upregulated, as seen in group 4. Regarding the heterogeneous nature of FGFR1 copy numbers in group 3, it is worthwhile investigating copy number‐dependent biological behaviour together with TME properties. TNBC samples in group 3 were further stratified based on CN < 5 and CN ≥5 of the FGFR1 average copy number (CN) [12, 14]. In group 3, patients with FGFR1 CN ≥ 5 exhibited significant enrichment of immune‐suppressive genes (e.g. IL6ST, FAS, CTNNA3, and DTX1), along with activation of pathways related to MAPK signalling, epithelial cell proliferation, and neutrophil migration. Their TME was predominantly characterized by peripheral T cells and resting mast cells. Conversely, patients with FGFR1 CN < 5 showed enrichment of immune‐activating genes (e.g. IL11) and pathways associated with immune activation, interferon‐γ response, T‐cell differentiation, and B‐cell activation. Their TME displayed higher infiltration of cytotoxic CD8 T cells and NK cells (Figure 4A–C).
Figure 4.

Variations in the tumour microenvironment across the group 3 subtype. (A) Volcano plot showing differentially expressed genes in group 3 (CN ≥ 5) compared with group 3 (CN < 5) patients (fold‐change > 1.5, p < 0.05). (B) Enrichment analysis of KEGG, Hallmark, Reactome, and GO terms of the DEGs upregulated in the group 3 (CN ≥ 5) and group 3 (CN < 5) groups. (C) Heatmap illustrating the normalized gene set variation analysis (GSVA) enrichment scores of various tumour microenvironment signatures across the group 3 (CN ≥ 5) and group 3 (CN < 5) groups.
Proposed criteria of FGFR1 amplification in TNBC
In group 3, patients were stratified based on FGFR1 CN, where a CN cutoff ≥ 5 identified a subset with molecular features resembling groups 1 and 2, characterized by enhanced pro‐oncogenic signalling pathway activation. For this study, tumours with FGFR1 amplified (FGFR1‐Amp) were defined as those meeting either of the following criteria: FGFR1/CEN8 ≥ 2.0 or CN ≥ 5.0. Tumours not fulfilling these thresholds were classified as FGFR1‐neutral (FGFR1‐Neu). Based on these criteria, we identified 26 tumours (26/182, 14.29%) with FGFR1 amplification (FGFR1‐Amp) and 156 cases without FGFR1 amplification (FGFR1‐Neu). There were no significant differences in the clinicopathological characteristics between the two groups (Table 2). Differential gene expression analysis revealed that FGFR1‐Amp tumours were enriched in tumour proliferation pathways, with significant upregulation of FGFR1 pathway molecules such as AKT1/2, ARAF, MTOR, PIK3CA, and FGF12/13 (Figure 5A,B). In contrast, FGFR1‐Neu tumours exhibited enrichment in immune activation pathways and higher immune cell infiltration (Figure 5A–C). Notably, IHC staining confirmed these findings, showing greater CD3+/CD8+ T‐cell infiltration in FGFR1‐Neu tumours compared with FGFR1‐Amp tumours (Figure 5D).
Table 2.
Comparison of the clinicopathological features in TNBCs with amplified or neutral FGFR1 FISH results. Data are presented as number (percentage) or median (range). Amplification is defined as FGFR1/CEN8 ratio ≥ 2.0 or FGFR1 copy number ≥ 5.0, while other cases are neutral.
| FGFR1 FISH | P1 | P2 | |||
|---|---|---|---|---|---|
| Amplified (n = 24) | Neutral (n = 154) | Neutral with MP results (n = 61) | |||
| Age (range), years | 51 (33–78) | 51.5 (26–87) | 47 (26–83) | 0.828 | 0.113 |
| Menopause | 0.678 | 0.371 | |||
| Pre‐ and peri‐ | 12 (50.0%) | 70 (45.5%) | 37 (60.7%) | ||
| Post‐ | 12 (50.0%) | 84 (54.5%) | 24 (39.3%) | ||
| AJCC stage | 0.218 | 0.096 | |||
| I | 12 (50.0%) | 65 (42.2%) | 19 (31.1%) | ||
| II | 6 (25.0%) | 68 (44.2%) | 32 (52.5%) | ||
| III | 5 (20.8%) | 18 (11.7%) | 8 (13.1%) | ||
| IV | 1 (4.2%) | 2 (1.3%) | 2 (3.3%) | ||
| Unknown | 0 (0) | 1 (0.6%) | 0 (0) | ||
| Histology* | 1.000 | 1.000 | |||
| IBC‐NST | 25 (96.2%) | 147 (94.2%) | 60 (96.8%) | ||
| Other | 1 (3.8%) | 9 (5.8%) | 2 (3.2%) | ||
| Histological grade* | 0.189 | 0.047 | |||
| Grade 1 | 1 (3.8%) | 3 (1.9%) | 0 (0) | ||
| Grade 2 | 3 (11.5%) | 36 (23.1%) | 12 (19.4%) | ||
| Grade 3 | 20 (76.9%) | 113 (72.4%) | 50 (80.6%) | ||
| Unknown | 2 (7.7%) | 4 (2.6%) | 0 (0) | ||
| HER2 expression* | 0.395 | 0.871 | |||
| Low | 10 (38.5%) | 74 (47.4%) | 25 (40.3%) | ||
| Zero | 16 (61.5%) | 82 (52.6%) | 37 (59.7%) | ||
| Ki‐67 index (range), %* | 70 (10–95) | 60 (5–90) | 70 (8–90) | 0.050 | 0.757 |
| Death | 1.000 | 0.928 | |||
| Present | 2 (8.3%) | 12 (7.8%) | 3 (4.9%) | ||
| Absent | 22 (91.7%) | 142 (92.2%) | 58 (95.1%) | ||
| Other malignancies | 0.237 | 0.415 | |||
| Present | 1 (4.2%) | 24 (15.6%) | 8 (13.1%) | ||
| Absent | 23 (95.8%) | 130 (84.4%) | 53 (86.9%) | ||
| Family history of malignancies | 0.390 | 0.585 | |||
| Breast cancer | 4 (16.7%) | 14 (9.1%) | 6 (9.8%) | ||
| Other malignancies | 4 (16.7%) | 22 (14.3%) | 9 (14.8%) | ||
| Absent | 16 (66.7%) | 118 (76.6%) | 46 (75.4%) | ||
P1: the p value between FGFR1‐amplified and FGFR1‐neutral cases; P2: the p value between FGFR1‐amplified cases and FGFR1‐neutral cases with MP results.
FISH, fluorescence in situ hybridization; MP, Master Panel; TNBC, triple‐negative breast cancer.
The majority of clinical data were analysed at the patient level, whereas tumour‐specific characteristics, including histological type, tumour grade, HER2 status, and Ki‐67 index, were assessed on a per‐tumour basis.
Figure 5.

FGFR1 amplified and its impact on the tumour microenvironment. (A) Enrichment analysis of KEGG, Hallmark, Reactome, and GO terms of the DEGs upregulated in the FGFR1‐Amp and FGFR1‐Neu groups. (B) Differentially expressed genes in the FGFR1 signalling pathway between the FGFR1‐Amp and FGFR1‐Neu groups. (C) Heatmap illustrating the normalized gene set variation analysis (GSVA) enrichment scores of various tumour microenvironment signatures across the FGFR1‐Amp and FGFR1‐Neu groups. (D) Quantification of IHC positive CD3+/CD4+/CD8 cells in the FGFR1‐Amp and FGFR1‐Neu groups. *p < 0.05, **p < 0.01, ***p < 0.001. p values of statistical significance are shown in bold.
FGFR1 and HER2 status synergistically delineated TNBC into biologically distinctive subcategories
Immune profiling revealed that FGFR1‐Amp & HER2‐low tumours exhibited the lowest stromal and immune scores and elevated tumour purity in samples (Figure 6A). Simultaneously, FGFR1‐Amp & HER2‐low tumours demonstrated increased tumour‐associated signalling activities, including epithelial–mesenchymal transition (EMT) and enhanced expression of endothelium and angiogenesis signatures, in comparison to the other subgroups (Figure 6B). Conversely, FGFR1‐Neu tumours, particularly those with HER2‐zero, showed increased immune cell infiltration. This involved NK cells, effector cells, MHC I/II molecules, and various other immune activation markers (Figure 6B). Remarkably, immune checkpoint markers (PDL1, CTLA4, HAVCR2, LAG3, PDCD1, and PDCD1LG2) were most prominently elevated in FGFR1‐Neu & HER2‐low/zero compared with FGFR1‐Amp & HER2‐low tumours (Figure 6C). GSEA analysis revealed that FGFR1‐Amp tumours were significantly enriched for DNA repair, kinase activity, and cell cycle pathways; TP53 binding was notably enriched in tumours characterized by both FGFR1 amplification and low HER2 expression (Figure 6D), while antigen processing pathways were significantly enriched in the FGFR1‐Neu tumours, particularly those with HER2‐zero.
Figure 6.

Immune profiles stratified by FGFR1 and HER2 status. (A) The analysis utilizes the ESTIMATE algorithm to quantify the immune and stromal scores, tumour purity, and overall ESTIMATE scores for each subtype. (B) Immune landscape and GSVA analysis across four distinct subtypes. (C) Expression levels of immune checkpoint molecules across the four different groups. (D) GSEA pathway enrichment analysis across the four groups. p values of statistical significance are shown in bold.
Discussion
In this study, the FGFR1 amplification status was proposed based on the FGFR1/CEN8 ratio and FGFR1 CN. FGFR1‐amplified tumours were defined as those with an FGFR1 ratio ≥ 2.0 (FGFR1/CEN8) or CN ≥ 5.0. Tumours that did not meet these criteria were considered FGFR1‐neutral. While Jafarian et al concluded that FGFR1 amplification lacked prognostic significance in TNBC [22], the distinct biological features between these two groups were supported by our transcriptomic data. In terms of FISH‐based detection, Mayo Clinic Laboratories defines FGFR1 as amplified when the FGFR1/CEN8 ratio exceeds 2.0 or when an average of six or more FGFR1 copies per tumour nucleus are observed. Tumours not meeting these criteria are deemed FGFR1‐negative [9]. However, other studies have used different thresholds for defining FGFR1 amplification. Luo et al defined FGFR1 amplification as an FGFR1 CN > 6 or an FGFR1/CEP8 ratio > 2.2 [23]. Furthermore, some studies have defined FGFR1 amplification as more than five signals per nucleus in more than 50% of cancer cells or when large gene copy clusters are present [12]. Previous studies have predominantly characterized FGFR1 amplification through centromere and absolute copy number metrics. Our research extends this framework by utilizing a multi‐omics methodology, which integrates DNA, transcriptomic, and IHC data, to delineate four distinct patient cohorts based on centromere and absolute CN status. Importantly, within the heterogeneous group 3, we further differentiated between oncogenic signalling activation and immune activation characteristics using an FGFR1 CN threshold of ≥ 5 or < 5. This integrative approach enabled us to establish a more precise molecular definition of FGFR1 amplification, thereby providing enhanced insights into the molecular mechanisms underpinning this genetic alteration. This advancement holds the potential to significantly enhance the precision of targeted therapies and improve prognosis prediction. Furthermore, the present study provides a comprehensive analysis of the impact of FGFR1 amplification on the tumour biology and immune microenvironment in TNBC. One of the most striking findings of our study is the differential immune microenvironment observed between FGFR1‐Amp and FGFR1‐Neu tumours. Our study demonstrated that FGFR1‐Neu tumours exhibited enhanced immune cell infiltration, characterized by elevated immune scores, increased immune cell invasion, and upregulation of immune checkpoint markers (CD274, HAVCR2, PDCD1LG2). This immune‐activated phenotype is consistent with previous findings that TNBC subtypes lacking FGFR1 amplification are more likely to exhibit a robust immune response [8, 24]. The elevated expression of CD3, CD4, and CD8 in FGFR1‐Neu groups further supports the notion that these tumours may be more responsive to immunotherapy.
In contrast, FGFR1‐Amp tumours were associated with a suppressive immune microenvironment and pro‐tumourigenic signatures. These findings are consistent with previous studies that have implicated FGFR1 amplification in promoting tumour growth and survival through activation of downstream signalling pathways such as MAPK and PI3K/AKT [2, 25]. Beyond its role in proliferative signalling, FGFR pathway aberrations have also been implicated in chemoresistance across breast cancer subtypes, including TNBC [26]. Given the association of FGFR1 amplification with an immunosuppressive TME, targeting this pathway represents a rational therapeutic strategy. FGFR inhibitors have been approved for specific solid tumours in recent years, but those targeting TNBC remain experimental. AZD4547 and BLU9931 exhibited efficacy in TNBC PDX/PDO models with enhanced FGFR expression [27]. Compound 6 suppresses TNBC migration and invasion by downregulating pFGFR1 and modulating EMT markers [28]. Clinically, André et al reported that dovitinib, a multi‐targeted FGFR inhibitor, showed antitumour activity in FGFR‐amplified breast cancer cell lines and may have activity in breast cancers with FGF pathway amplification [5]. Mechanistically, FGFR signalling reshapes the tumour microenvironment to facilitate immune escape [29]. Wu et al showed that FGFR blockade disrupts cancer‐associated fibroblast (CAF) function, reduces VCAM‐1‐mediated physical barriers, and enhances CD8+ T‐cell infiltration in TNBC [3]. Ruan et al further summarized that FGFR signalling upregulates immunosuppressive cytokines (e.g. TGF‐β) and angiogenic factors (e.g. VEGF), making FGFR inhibition a promising strategy to reverse TME immunosuppression in TNBC [8].
Targeting FGFR1 signalling pathways, either through FGFR inhibitors or combination therapies that include immune modulators, may be a viable approach for these patients [30, 31]. While immune checkpoint blockade (ICB) monotherapy has shown controversial efficacy in FGFR‐altered tumours, emerging preclinical and clinical evidence supports the superior antitumour activity of combining FGFR tyrosine kinase inhibitors (FGFR‐TKIs) with ICB compared with either therapy alone [32, 33]. For instance, lenvatinib, a multi‐targeted TKI, synergizes with anti‐PD‐1 therapy by remodelling the immunosuppressive tumour microenvironment, promoting tumour vascular normalization, and amplifying cytotoxic T‐cell activity [34]. This combinatorial approach not only enhances therapeutic efficacy but also induces durable immune memory, as demonstrated in hepatocellular carcinoma models and clinical trials [32]. Such synergy highlights the potential of dual targeting of FGFR signalling and immune checkpoints to overcome resistance mechanisms in FGFR‐driven malignancies.
Prior research has indicated that HER2‐low tumours may possess distinct biological characteristics compared with HER2‐zero tumours [35, 36]. Single‐cell RNA‐sequencing analysis showed that HER2‐low TNBC is more metabolically active and aggressive, whereas HER2‐zero TNBC also demonstrated a more active immune microenvironment, with enhanced macrophage polarization, abundant CD8+ T cells, diverse T‐cell receptors, and higher levels of immunotherapy markers, contributing to a better immunotherapeutic response [35]. Our findings underscore a significant interaction between FGFR1 and HER2 in TNBC. Specifically, tumours exhibiting both FGFR1 amplification and HER2 expression displayed the most immunosuppressive phenotype. FGFR1‐Amp & HER2‐low tumours showed heightened tumour signalling, including EMT and increased endothelium and angiogenesis markers. Considering the substantial advancements in HER2‐targeted antibody–drug conjugates (ADCs) for breast cancer treatment, especially their notable clinical efficacy in populations with low HER2 expression [37], the therapeutic strategy of combining agents targeting FGFR1 inhibitors with ADCs may constitute a novel and promising avenue for research in patients with FGFR1‐Amp/ HER2‐low TNBC. In contrast, FGFR1‐Neu tumours, especially HER2‐zero, exhibited greater immune cell infiltration, featuring NK cells, effector cells, MHC I/II molecules, and other immune activation indicators. This immunologically ‘hot’ tumour microenvironment suggests that this particular subgroup may derive greater benefit from immunotherapy [38]. Collectively, these findings provide a useful way to group TNBC patients based on FGFR1 CN and HER2 levels. This grouping suggests that patients with FGFR1‐Neu/HER2‐zero tumours may benefit more from immunotherapy, while those with FGFR1‐Amp/HER2‐low tumours could be candidates for treatments that combine FGFR inhibitors with other agents.
This study has several limitations that should be acknowledged. First, the retrospective design and the relatively small sample size of FGFR1‐Amp tumours may limit the generalizability of our findings. While we observed significant differences in immune activation between FGFR1 subtypes, these results need to be validated in larger, prospective studies. Second, although our study provides comprehensive transcriptomic and immune profiling data, the causal relationship between FGFR1 amplification and immune suppression remains to be established. Functional validation through in vitro and in vivo models is essential to confirm whether FGFR1 amplification directly drives immune evasion. Third, the lack of robust follow‐up data prevented a definitive analysis of the prognostic value of FGFR1 amplification. A prospectively monitored cohort is needed for future validation. Finally, given that the FGFR1 FISH results were based on TMA samples, cases with signals near the amplification cutoff should be interpreted with caution. In clinical practice, analysis on whole‐tissue sections may be preferable to mitigate potential bias from intratumoural heterogeneity.
Author contributions statement
YG conceived experiments, collected samples and analysed data. FY analysed data. JP and TS carried out experiments. YG, FY, CZ, SW and ZL interpreted the data. YG drafted the manuscript. All authors have read and approved the final paper.
Supporting information
Figure S1. TNBC subtype distribution across the FGFR1 copy number subgroups
Figure S2. Correlation between FGFR1 copy numbers (CN) detected by fluorescence in situ hybridization (FISH) and next‐generation sequencing (NGS) across the FGFR1 subgroups
Table S1. Immunohistochemical conditions and antibodies used
Table S2. List of genes detected for single nucleotide variant (SNV) and insertion/deletion (Indel) analysis in the AmoyDx® Master Panel
Table S3. List of genes detected for copy number variation (CNV) analysis in the AmoyDx® Master Panel
Table S4. List of genes analysed for fusion and alternative splicing events in the AmoyDx® Master Panel
Table S5. List of genes assessed for RNA expression in the AmoyDx® Master Panel
Table S6. Clinicopathological characteristics of TNBC patients with different copy number subgroups (n = 178)
Acknowledgements
The authors received no specific funding for this work.
No conflicts of interest were declared.
Contributor Information
Shafei Wu, Email: shafei_wu@126.com.
Zhiyong Liang, Email: liangzy@pumch.cn.
Data availability statement
The data generated in this study are available within the article and its supplementary data files. Further supporting data are available from the corresponding author upon reasonable request.
References
- 1. Lin H, Hassan Safdar M, Washburn S, et al. Fibroblast growth receptor 1 is regulated by G‐quadruplex in metastatic breast cancer. Commun Biol 2024; 7 : 963. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Turner N, Grose R. Fibroblast growth factor signalling: from development to cancer. Nat Rev Cancer 2010; 10 : 116–129. [DOI] [PubMed] [Google Scholar]
- 3. Wu Y, Yi Z, Li J, et al. FGFR blockade boosts T cell infiltration into triple‐negative breast cancer by regulating cancer‐associated fibroblasts. Theranostics 2022; 12 : 4564–4580. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Gu W, Yang J, Wang Y, et al. Comprehensive identification of FGFR1‐4 alterations in 5 557 Chinese patients with solid tumors by next‐generation sequencing. Am J Cancer Res 2021; 11 : 3893–3906. [PMC free article] [PubMed] [Google Scholar]
- 5. André F, Bachelot T, Campone M, et al. Targeting FGFR with dovitinib (TKI258): preclinical and clinical data in breast cancer. Clin Cancer Res 2013; 19 : 3693–3702. [DOI] [PubMed] [Google Scholar]
- 6. Turner N, Pearson A, Sharpe R, et al. FGFR1 amplification drives endocrine therapy resistance and is a therapeutic target in breast cancer. Cancer Res 2010; 70 : 2085–2094. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Hu Y, Lu Y, Xing F, et al. FGFR1/MAPK‐directed brachyury activation drives PD‐L1‐mediated immune evasion to promote lung cancer progression. Cancer Lett 2022; 547 : 215867. [DOI] [PubMed] [Google Scholar]
- 8. Ruan R, Li L, Li X. Unleashing the potential of combining FGFR inhibitor and immune checkpoint blockade for FGF/FGFR signaling in tumor microenvironment. Mol Cancer 2023; 22 : 60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. FGFR1 (8p11.2) Amplification, FISH, Tissue [Internet]. Mayo Clinic Laboratories. Test ID: FGF1F. [Accessed August 27, 2026]: Available from: https://www.mayocliniclabs.com/test-catalog/Overview/58124#Overview.
- 10. Jang M, Kim E, Choi Y, et al. FGFR1 is amplified during the progression of in situ to invasive breast carcinoma. Breast Cancer Res 2012; 14 : R115. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Brunello E, Brunelli M, Bogina G, et al. FGFR‐1 amplification in metastatic lymph‐nodal and haematogenous lobular breast carcinoma. J Exp Clin Cancer Res 2012; 31 : 103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Elbauomy Elsheikh S, Green AR, Lambros MB, et al. FGFR1 amplification in breast carcinomas: a chromogenic in situ hybridisation analysis. Breast Cancer Res 2007; 9 : R23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Wolff AC, Somerfield MR, Dowsett M, et al. Human epidermal growth factor receptor 2 testing in breast cancer: ASCO–College of American Pathologists guideline update. J Clin Oncol 2023; 41 : 3867–3872. [DOI] [PubMed] [Google Scholar]
- 14. Gonzalez‐Ericsson PI, Servetto A, Formisano L, et al. FGFR1 antibody validation and characterization of FGFR1 protein expression in ER+ breast cancer. Appl Immunohistochem Mol Morphol 2022; 30 : 600–608. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Zhao S, Ma D, Xiao Y, et al. Molecular subtyping of triple‐negative breast cancers by immunohistochemistry: molecular basis and clinical relevance. Oncologist 2020; 25 : e1481–e1491. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Jiang YZ, Ma D, Suo C, et al. Genomic and transcriptomic landscape of triple‐negative breast cancers: subtypes and treatment strategies. Cancer Cell 2019; 35 : 428–440.e5. [DOI] [PubMed] [Google Scholar]
- 17. Gulhan DC, Lee JJ, Melloni G, et al. Detecting the mutational signature of homologous recombination deficiency in clinical samples. Nat Genet 2019; 51 : 912–919. [DOI] [PubMed] [Google Scholar]
- 18. Bagaev A, Kotlov N, Nomie K, et al. Conserved pan‐cancer microenvironment subtypes predict response to immunotherapy. Cancer Cell 2021; 39 : 845–865.e7. [DOI] [PubMed] [Google Scholar]
- 19. Hänzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA‐seq data. BMC Bioinformatics 2013; 14 : 7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Newman AM, Liu CL, Green MR, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods 2015; 12 : 453–457. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Becht E, Giraldo NA, Lacroix L, et al. Estimating the population abundance of tissue‐infiltrating immune and stromal cell populations using gene expression. Genome Biol 2016; 17 : 218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Jafarian AH, Kooshkiforooshani M, Farzad F, et al. The relationship between fibroblastic growth factor receptor‐1 (FGFR1) gene amplification in triple negative breast carcinomas and clinicopathological prognostic factors. Iran J Pathol 2019; 14 : 299–304. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Luo J, Liu S, Leung S, et al. An mRNA gene expression‐based signature to identify FGFR1‐amplified estrogen receptor‐positive breast tumors. J Mol Diagn 2017; 19 : 147–161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Emens LA. Breast cancer immunotherapy: facts and hopes. Clin Cancer Res 2018; 24 : 511–520. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. André F, Cortés J. Rationale for targeting fibroblast growth factor receptor signaling in breast cancer. Breast Cancer Res Treat 2015; 150 : 1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Chen SH, Tse KP, Lu YJ, et al. Comprehensive genomic profiling and therapeutic implications for Taiwanese patients with treatment‐naïve breast cancer. Cancer Med 2024; 13 : e7384. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Chew NJ, Lim Kam Sian T, Nguyen EV. Evaluation of FGFR targeting in breast cancer through interrogation of patient‐derived models. Breast Cancer Res 2021; 23 : 82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Wang Y, Shen Z, Chen R. Discovery and characterization of novel FGFR1 inhibitors in triple‐negative breast cancer via hybrid virtual screening and molecular dynamics simulations. Bioorg Chem 2024; 150 : 107553. [DOI] [PubMed] [Google Scholar]
- 29. Tsimafeyeu I. Future perspectives: targeting fibroblast growth factor receptor 1 to enhance the efficacy of immunotherapy. Explor Target Antitumor Ther 2025; 6 : 1002327. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Dienstmann R, Rodon J, Prat A. Genomic aberrations in the FGFR pathway: opportunities for targeted therapies in solid tumors. Ann Oncol 2014; 25 : 552–563. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Nogova L, Sequist LV, Perez Garcia JM. Evaluation of BGJ398, a fibroblast growth factor receptor 1‐3 kinase inhibitor, in patients with advanced solid tumors harboring genetic alterations in fibroblast growth factor receptors: results of a global phase I, dose‐escalation and dose‐expansion study. J Clin Oncol 2017; 35 : 157–165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Deng H, Kan A, Lyu N. Dual vascular endothelial growth factor receptor and fibroblast growth factor receptor inhibition elicits antitumor immunity and enhances programmed cell death‐1 checkpoint blockade in hepatocellular carcinoma. Liver Cancer (Basel) 2020; 9 : 338–357. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Palakurthi S, Kuraguchi M, Zacharek SJ. The combined effect of FGFR inhibition and PD‐1 blockade promotes tumor‐intrinsic induction of antitumor immunity. Cancer Immunol Res 2019; 7 : 1457–1471. [DOI] [PubMed] [Google Scholar]
- 34. Kato Y, Tabata K, Kimura T. Lenvatinib plus anti‐PD‐1 antibody combination treatment activates CD8+ T cells through reduction of tumor‐associated macrophage and activation of the interferon pathway. PLoS One 2019; 14 : e0212513. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Hu XE, Yang P, Chen S, et al. Clinical and biological heterogeneities in triple‐negative breast cancer reveals a non‐negligible role of HER2‐low. Breast Cancer Res 2023; 25 : 34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Vos H, Lambein K, Richard F. Comparison of the tumor immune microenvironment of primary hormone receptor‐negative HER2‐positive and triple negative breast cancer. NPJ Breast Cancer 2021; 7 : 128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Modi S, Jacot W, Yamashita T, et al. Trastuzumab deruxtecan in previously treated HER2‐low advanced breast cancer. N Engl J Med 2022; 387 : 9–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Nolan E, Lindeman GJ, Visvader JE. Deciphering breast cancer: from biology to the clinic. Cell 2023; 186 : 1708–1728. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1. TNBC subtype distribution across the FGFR1 copy number subgroups
Figure S2. Correlation between FGFR1 copy numbers (CN) detected by fluorescence in situ hybridization (FISH) and next‐generation sequencing (NGS) across the FGFR1 subgroups
Table S1. Immunohistochemical conditions and antibodies used
Table S2. List of genes detected for single nucleotide variant (SNV) and insertion/deletion (Indel) analysis in the AmoyDx® Master Panel
Table S3. List of genes detected for copy number variation (CNV) analysis in the AmoyDx® Master Panel
Table S4. List of genes analysed for fusion and alternative splicing events in the AmoyDx® Master Panel
Table S5. List of genes assessed for RNA expression in the AmoyDx® Master Panel
Table S6. Clinicopathological characteristics of TNBC patients with different copy number subgroups (n = 178)
Data Availability Statement
The data generated in this study are available within the article and its supplementary data files. Further supporting data are available from the corresponding author upon reasonable request.
