A study on the mechanism of action of B7H4 in HER2-positive gastric cancer and its sensitivity to trastuzumab therapy Authors: Qiangzu Shao, Fan Zhang, Wenjie Wang, Zeping Huang Published: 2026 ABSTRACT BackgroundHuman epidermal growth factor receptor 2 (HER2)-positive gastric cancer (GC) is characterized by high malignancy and a poor prognosis. Trastuzumab is the first-line targeted therapy for this disease, but the frequent development of primary and acquired resistance severely compromises treatment efficacy and impedes improvements in patient outcomes. B7H4 is a critical negative immune checkpoint molecule that has been demonstrated to contribute to tumor progression and targeted therapy resistance in multiple cancers. However, the specific mechanisms by which B7H4 regulates sensitivity to trastuzumab in HER2-positive GC remain unclear. This study aims to investigate the expression and biological functions of B7H4 in HER2-positive GC, elucidate the molecular mechanisms underlying B7H4-mediated trastuzumab resistance, and thereby provide a theoretical basis for targeted resistance intervention and therapeutic optimization for this malignancy.MethodsThe study began with an analysis of The Cancer Genome Atlas (TCGA) database and immunohistochemical staining of tissue sections to assess the expression levels of B7H4 and HER2, as well as the correlation between their expression. Furthermore, the correlation between B7H4 expression and various pathological parameters in patients with HER2-positive GC was analysed. Western blot analysis was used to screen for co-expressing cells, and short hairpin RNA (shRNA) was employed to knockdown B7H4. The effects of B7H4 knockdown on the proliferation, migration and invasion of HER2-positive GC cells were assessed using colony formation assays, Cell Counting Kit-8 (CCK-8) assays, cell scratch assays and Transwell assays, respectively. RNA sequencing (RNA-Seq) was utilised to analyse the biological processes and signalling pathways regulated by B7H4 and to detect relevant biomarkers. Colony formation assays and CCK-8 assays were employed to evaluate the therapeutic sensitivity of trastuzumab to HER2-positive GC cells following B7H4 knockdown.ResultsAnalysis of the TCGA database and immunohistochemical staining of tissue sections revealed that B7H4 and HER2 were co-expressed in GC tissues, and their expression levels were positively correlated. Clinical correlation analysis revealed that B7H4 expression was significantly associated with tumor size, grade, depth of invasion, lymph node metastasis, distant metastasis, vascular invasion and nerve invasion in HER2-positive GC patients. Western blot analysis demonstrated co-expression of B7H4 and HER2 in NCI-N87 cells. Knockdown of B7H4 resulted in varying degrees of inhibition of proliferation, migration and invasion in NCI-N87 cells. RNA-seq results indicated that B7H4 knockdown affected biological processes such as cell proliferation, migration, invasion and epithelial-mesenchymal transition (EMT), and was significantly associated with the nuclear factor κB (NF-κB) signalling pathway. Knockdown of B7H4 significantly enhanced the sensitivity of HER2-positive GC cells to trastuzumab.ConclusionsB7H4 is highly expressed in HER2-positive GC and is associated with poor prognosis. B7H4 promotes tumor cell proliferation, migration and invasion by activating the NF-κB signalling pathway and driving the EMT process. B7H4 expression influences the sensitivity of HER2-positive GC cells to trastuzumab; inhibition of B7H4 significantly enhances the antitumor efficacy of trastuzumab. Highlight box Key findings • B7H4 is significantly co‑expressed with human epidermal growth factor receptor 2 (HER2) in gastric cancer (GC) tissues, and its high expression correlates with tumor size, grade, depth of invasion, lymph node metastasis, distant metastasis, and vascular/neural invasion. • Knockdown of B7H4 suppresses proliferation, migration, and invasion of HER2‑positive GC cells. Mechanistically, B7H4 activates the NF‑κB signalling pathway and drives epithelial-mesenchymal transition (EMT). • B7H4 knockdown markedly enhances trastuzumab sensitivity, reducing the half-maximal inhibitory concentration (IC50) by approximately 4.4‑fold. What is known and what is new? • HER2-positive GC is aggressive and prone to trastuzumab resistance; B7H4 is an immune checkpoint molecule that promotes tumor progression and immune evasion in various cancers, but its role in HER2-positive GC is unclear. • This study is the first to demonstrate that B7H4 is not only co-expressed with HER2 but also directly promotes malignancy in HER2-positive GC via the nuclear factor κB (NF-κB)/EMT axis. Moreover, we reveal that B7H4 modulates trastuzumab sensitivity, providing a mechanistic link between B7H4 and drug resistance. What is the implication, and what should change now? • B7H4 may serve as a prognostic biomarker and a novel therapeutic target. Combining B7H4 inhibition with trastuzumab could be a promising strategy to overcome resistance and improve clinical outcomes in HER2-positive GC. • Further validation in multiple cell lines and in vivo models, followed by clinical trials evaluating anti-B7H4 agents combined with trastuzumab in patients, are warranted to translate these findings into practice. Introduction According to the 2022 global cancer statistics provided by GLOBOCAN, gastric cancer (GC) is the fifth leading cause of cancer-related deaths worldwide and the fifth most common malignant tumor (1). In China, both the incidence and mortality rates of GC rank among the top three nationally; the 5-year survival rate stands at only 31.9%, which is significantly lower than that of other malignant tumors, making it a major challenge threatening the health of the Chinese population (2). In China, 12–13% of GC cases are human epidermal growth factor receptor 2 (HER2)-positive. Compared with standard GC, patients with HER2-positive GC present at a later clinical stage, exhibit greater tumor invasiveness, higher rates of metastasis, and a poorer prognosis (3). Although the clinical outcomes for patients with HER2-positive GC have improved following targeted HER2 therapy, some patients still experience issues such as low response rates, significant adverse effects, and treatment resistance (4). Therefore, it is crucial to conduct in-depth research into the target genes and molecular mechanisms underlying the development and progression of HER2-positive GC, and to seek new and effective combination treatment regimens. The B7 family of immune checkpoints is associated with the development of various cancers; among these, research on programmed cell death protein 1/programmed cell death-ligand 1 (PD-1/PD-L1) has been the most extensive and has brought about a revolutionary transformation in cancer treatment. B7-H4 is a member of the B7 family, also known as B7x, B7S1, and VTCN1, and was first identified in 2003 (5). Recent research has significantly expanded our understanding of B7-H4, revealing its multifaceted involvement in various malignant tumors (6). Studies have shown that B7H4 is a single-pass transmembrane glycoprotein that inhibits T-cell function via a receptor whose identity remains to be determined. It is expressed in various cancers (such as breast cancer, ovarian cancer, and GC), and its degradation process can be regulated by post-translational modifications. It promotes the formation of an immunosuppressive tumor microenvironment, mediates tumor immune evasion, and facilitates the malignant biological behaviour of tumor cells; furthermore, its expression levels are closely associated with patient survival prognosis (7-10). However, the role and molecular mechanisms of B7H4 in HER2-positive GC remain unclear. This study aims to investigate the effects of B7H4 on the biological behaviour of HER2-positive GC cells and elucidate its underlying mechanisms of action. We present this article in accordance with the MDAR reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2026-1121/rc). Methods Clinical data and tissue sample sources All clinical data and tissue samples used in this study were obtained from The Second Hospital & Clinical Medical School. All patients were diagnosed with HER2-positive gastric adenocarcinoma based on histopathological examination and had not received any treatment prior to surgery. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of The Second Hospital & Clinical Medical School (No. 2021A-314). Informed consent was taken from all the patients. Cells, reagents and equipment The human GC cell lines NCI-N87, MKN74, HGC27, AGS and normal human gastric epithelial cells (GES) were all purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). DMEM, RPMI1640, foetal bovine serum (FBS), penicillin-streptomycin solution and 0.25% trypsin-EDTA (1×) used for cell culture were all purchased from Gibco (Shanghai, China). The immunohistochemistry staining kit was purchased from Beijing Zhongshan Jinqiao Co., Ltd. (Beijing, China), whilst the Cell Counting Kit-8 (CCK-8) assay kit, Trizol and radio immunoprecipitation assay (RIPA) lysis buffer were purchased from China Biyuntian Co., Ltd. (Shanghai, China), the bicinchoninic acid (BCA) protein concentration assay kit and PVDF membranes were purchased from Thermo Fisher Scientific (Shanghai, China), the RNA reverse transcription and qPCR kits were purchased from Suzhou Boaolong Technology Co., Ltd. (Suzhou, China), Transwell chambers and Matrigel matrix gel were purchased from Beijing Biyuntian Co., Ltd. (Beijing, China), the primary and secondary antibodies used in the experiments were purchased from Suzhou Boaolong Technology Co., Ltd., and transfection reagents and lentiviruses were purchased from Shanghai Jima Gene Co., Ltd. (Shanghai, China). Experimental methods Cell culture Human GC cells and GES were cultured in complete RPMI-1640 medium containing 10% FBS and 1% penicillin-streptomycin. The incubator was maintained at 37 ℃ with a CO2 concentration of 5%. Cells were passaged when confluence reached 80%. Lentiviral transfection and screening The lentiviruses used for transfection were supplied by Shanghai Jima Pharmaceutical Technology Co., Ltd. Transfection was performed using 6-well plates, with the following groups: Blank group, sh-B7H4#NC group, sh-B7H4#1 group, sh-B7H4#2 group, and sh-B7H4#3 group. After seeding NCI-N87 cells into the plates, an appropriate volume of complete medium was added, and the plates were incubated overnight at 37 ℃ and 5% CO2; after 24 hours, when cell confluence reached 50%, the medium was replaced and the cells were washed. A mixture of 1.8 mL RPMI-1640, 200 µL viral stock and 2 µL polybrene was thoroughly mixed and added to the cells; after 48 hours of culture, the medium was replaced with complete medium, and cells were selected using an appropriate concentration of puromycin to obtain stably transfected cells. CCK-8 assay Seed experimental and control group cells (5,000 cells per well) into a 96-well plate for culture. After culturing the cells in each group for 6, 24, 48, and 72 hours, add 10 µL of CCK-8 reagent per well during the final hour of culture. Determine cell proliferation by measuring the absorbance (at 450 nm) of each well. Colony formation assay Digested cells from the experimental and control groups were seeded into 6-well plates (3,000–5,000 cells per well) and cultured in an incubator until colonies were visible to the naked eye. The culture medium was discarded, and the wells were washed 2–3 times with phosphate buffered saline (PBS). The cells were then fixed with 4% paraformaldehyde for 10–15 minutes, stained with crystal violet for 10–15 minutes, rinsed and air-dried, and images were captured and counted using ImageJ software. Cell scratch assay For the cell scratch assay, several parallel horizontal lines were drawn on the bottom of a 6-well plate at equal intervals using a marker pen and a ruler. NCI-N87 cells were digested and counted, with 5×104 cells seeded into each well of the 6-well plate, which was then placed in a cell culture incubator for cultivation. When cell confluence reaches 90–95% or higher, use a 200 µL sterile, enzyme-free pipette tip to rapidly draw a straight line perpendicular to the bottom horizontal lines. Wash twice with PBS to remove detached cells, and photograph the cell scratch pattern at this stage. Add 2 mL of serum-free medium and continue culture for 24 hours; photograph the changes in cell migration and analyse the migration patterns of tumor cells in different groups using ImageJ software. Transwell assay Add 50 µL of serum-free medium to a 24-well plate. Place the Transwell units into the wells and incubate in a 37 ℃ incubator for 30 minutes to hydrate the basement membrane. Take 100 µL of the matrix gel diluted in the specified ratio (serum-free medium 1640:matrix gel =1:8) and evenly coat the upper chamber of the Transwell unit. Place in the incubator for 4–6 hours to allow the matrix gel to solidify. Count the cells to be treated. Resuspend the cells in 200 µL of empty medium and add to the upper chamber, ensuring 2×105 cells are seeded in each well. Add 500 µL of complete medium to the lower chamber and incubate at 37 ℃ for 48 hours. Place the chambers in an empty 24-well plate, wash the chambers three times with PBS for 3 minutes each, then add 4% paraformaldehyde to fix for 10–15 minutes. Wash the chambers three times with PBS for 3 minutes each, then stain with crystal violet for 10–15 minutes. Rinse thoroughly with a gentle stream of tap water, then wipe away excess matrix gel and cells from the upper chamber using a cotton swab, and leave to air-dry at room temperature. Take photographs under a microscope and count the number of cells using ImageJ software. Quantitative real-time polymerase chain reaction (qRT-PCR) Total RNA was extracted from cell and tissue samples using Trizol reagent, and reverse transcribed into cDNA using PrimeScript reverse transcriptase according to the manufacturer’s instructions. qRT-PCR was performed using specific primers, with the GAPDH mRNA transcript as a control, to quantify the relative expression levels of the target gene. Primer sequences: B7H4: forward 5'-CATCATCATTATTCTGGCTGGAG-3', reverse 5'-AGGCGACAGTAGTGACTGTGATGG-3'; GAPDH: forward 5'-GAAGGTGAAGGTCGGAGTC-3', reverse 5'-GAGATGGTGATGGGATTTC-3'. The relative expression levels of B7H4 were analysed using the 2−ΔΔCt method. Western blotting Proteins were extracted from cells using RIPA lysis buffer containing protease and phosphatase inhibitors, and quantified using a BCA protein assay kit. Equal amounts (50 µg) of protein were separated on a 10–15% polyacrylamide gel via sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a polyvinylidene fluoride (PVDF) membrane. The membrane was blocked with 5% skimmed milk, incubated with the primary antibody overnight at 4 ℃. On the following day, the membrane was washed with a Tween-20 triethanolamine buffered saline solution, then incubated with a horseradish peroxidase-conjugated secondary antibody. Bands were visualised using an enhanced chemiluminescence (ECL) kit and analysed using ImageJ software. Immunohistochemistry (IHC) IHC was employed to detect the expression of HER2 and B7H4 in tumor tissues from 50 patients with GC. Paraffin-embedded tissue sections were deparaffinised, rehydrated, antigen-recovered and blocked, followed by the addition of primary antibodies (rabbit anti-human HER2 monoclonal antibody and rabbit anti-human B7H4 monoclonal antibody, diluted 1:200 by volume) and incubated overnight at 4 ℃. On the following day, secondary antibodies were added, and colouring was performed using 3,3’-diaminobenzidine (DAB) as the chromogen. The sections were counterstained with haematoxylin, dehydrated, and mounted with neutral resin. Scoring is based on the percentage of stained cells (0 represents 0–5%, 1 represents 6–25%, 2 represents 26–50%, 3 represents 51–75%, 4 represents 76–100%) and staining intensity (0 indicates no staining, 1 indicates weak staining, 2 indicates moderate staining, 3 indicates strong staining). The final B7H4 and HER2 expression scores for the tissue sections are calculated by multiplying the two results. A total score of 0 is negative, 1–4 is weakly positive (+), 5–6 is moderately positive (++), and 6 or above is strongly positive (+++). For the purposes of analysis, samples with a score of 4 or less are classified as the low-expression group, and samples with a score of 5–9 are classified as the high-expression group. RNA sequencing (RNA-seq) and bioinformatics analysis The B7H4 gene was knocked down in NCI-N87 cells. Following subculturing and a 3-day selection with puromycin, total RNA was extracted using TRIzol and sent to Shanghai Qingke Biotechnology Co., Ltd. (Shanghai, China) for sequencing. The sequencing platform used was the Illumina NovaSeq, and the Illumina Truseq™ RNA Sample Prep Kit was employed for library preparation. Following sequencing, gene expression was quantified using RNA-Seq by Expectation-Maximization (RSEM) software, with gene expression levels calculated based on read counts. Differentially expressed genes were analysed using DESeq2 software, with the criteria for significance defined as false discovery rate (FDR) <0.05 and |log2fold change (FC)| ≥1. Subsequently, bioinformatic analyses including Gene Ontology (GO) analysis and gene set enrichment analysis (GSEA) were performed on the differentially expressed genes. Statistical analysis Statistical analysis and plotting of experimental data were performed using GraphPad Prism 10.2 software. Quantitative data are expressed as mean ± standard deviation. Data analysis between two groups was performed using the Student’s t-test, whilst analysis involving three or more groups was performed using one-way analysis of variance (ANOVA). Following the identification of significant differences, post-hoc tests were conducted using the Student-Newman-Keuls method to determine the groups exhibiting significant differences. For the comparison of B7H4 protein expression differences in HER2-positive GC tissues based on IHC results, an independent samples t-test was employed. The association between B7H4 expression differences and clinical TNM staging was assessed using the chi-square test. Spearman’s rank correlation test was used for correlation analysis. In the analysis of differentially expressed genes, the P value represents the significance level, calculated using a negative binomial distribution. In gene enrichment analysis, the P value represents the significance level, calculated based on the hypergeometric distribution. A P value <0.05 was considered statistically significant. Results Expression of B7H4 and HER2 in GC This study first explored the correlation between B7H4 and HER2 expression in tumor tissues based on transcriptomic data from GC tissues in The Cancer Genome Atlas (TCGA) public database. The results showed that the mRNA expression of B7H4 and HER2 in GC tissues was positively correlated, with the difference being statistically significant (Figure 1A,1B). Analysis of the expression and correlation between B7H4 and HER2 in gastric cancer, based on the TCGA database and clinical pathology slides. (A,B) Analysis of the expression and correlation between B7H4 and HER2 in the TCGA database. (C,D) Representative IHC images of gastric cancer tumor tissue showing B7H4 and HER2 expression with DAB staining and hematoxylin counterstaining. Analysis of the expression and correlation between B7H4 and HER2 in clinical pathology slides. HER2, human epidermal growth factor receptor 2; IHC, immunohistochemistry; TCGA, The Cancer Genome Atlas. To further validate the correlation between B7H4 and HER2 protein expression in GC, this study collected postoperative paraffin-embedded tissue specimens from 50 HER2-positive GC patients at the Second Hospital of Lanzhou University and simultaneously assessed the expression of B7H4 and HER2 proteins using the IHC method. The results indicated that B7H4 and HER2 co-expressed in GC tissues (Figure 1D). Following a paired correlation analysis of B7H4 and HER2 expression, the results showed a positive correlation between the two (Figure 1C), which was highly consistent with the transcriptomic analysis results from the TCGA database. These findings jointly confirm, at both the genetic and protein levels, a trend towards co-expression of B7H4 and HER2 in GC tissues, providing important clinical evidence and a theoretical basis for further in-depth research into the biological function and clinical significance of B7H4 in HER2-positive GC. Analysis of the association between B7H4 expression and clinical and pathological factors in patients To further clarify the clinical significance of B7H4 expression, this study conducted a systematic analysis of the clinical data and B7H4 expression status of the 50 enrolled HER2-positive GC patients. The variables examined included patient age, sex, tumor size, histological grade, Lauren classification, primary tumor depth of invasion (T stage), regional lymph node metastasis status (N stage), distant metastasis status (M stage), pathological TNM staging (pTNM staging), and the presence or absence of vascular and nerve invasion. Statistical analysis revealed (Table 1) that among the 50 HER2-positive GC patients, 24 were B7H4-positive (48%) and 26 were B7H4-negative (52%). The expression level of the B7H4 protein was significantly positively correlated with several key pathological indicators reflecting tumor invasiveness and metastatic potential. Specifically, high B7H4 expression was significantly associated with tumor size (P=0.01), histological grade (P=0.01), depth of tumor invasion (T stage) (P=0.02), lymph node metastasis (N stage) (P=0.02) and the incidence of distant metastasis (M stage) (P=0.046). Furthermore, high B7H4 expression was significantly associated with vascular invasion (P=0.01) and nerve invasion (P=0.005). These results consistently indicate that upregulated B7H4 expression is closely associated with the more aggressive biological behaviour of HER2-positive GC. On the other hand, in this study cohort, the expression status of B7H4 did not show a statistically significant association with patient age, sex, Lauren classification, or pTNM staging (P>0.05). The above results suggest that B7H4 is involved in specific malignant progression processes, such as local invasion and lymph node metastasis, in HER2-positive GC, and may serve as a potential therapeutic target. Table 1Correlation between B7H4 expression in HER2-positive gastric cancer and patients’ clinical and pathological characteristicsCharacteristicsB7H4 IHC scorePPositive (n=24)Negative (n=26)Age, yearsNA ≤601010 >601416GenderNA Male1821 Female65Tumor length, cm0.01 ≤5515 >51911Tumor differentiation0.01 Poorly156 Moderately713 Well27Lauren classificationNA Intestinal type1517 Diffuse type66 Mixed type33Clinical T stage0.02 T1 + T2818 T3 + T4168Clinical N stage0.01 N0616 N11810Clinical M stage0.046 M02026 M140pTNM stageNA I + II715 III + IV1711Vascular invasion0.005 Yes1910 No516Nerve invasion0.01 Yes1810 No616HER2, human epidermal growth factor receptor 2; IHC, immunohistochemistry; M, metastasis; N, node; NA, not applicable; pTNM, pathological tumor-node-metastasis; T, tumor. Correlation between B7H4 expression in HER2-positive gastric cancer and patients’ clinical and pathological characteristics HER2, human epidermal growth factor receptor 2; IHC, immunohistochemistry; M, metastasis; N, node; NA, not applicable; pTNM, pathological tumor-node-metastasis; T, tumor. Differences in B7H4 and HER2 expression in normal gastric mucosal epithelial cells and GC cells Preliminary Western blot experiments were conducted to examine the expression levels of B7H4 and HER2 in the GC cell lines NCI-N87, MKN74, AGS, and HGC-27, and the normal gastric mucosal epithelial cell line GES. The results showed that B7H4 expression was significantly higher in the GC cell lines NCI-N87 and MKN74 than in the other cell lines, whilst HER2 was highly expressed only in the NCI-N87 cell line and was not expressed in the other cell lines (Figure 2A). Based on these results, NCI-N87 cells were selected for subsequent functional studies. Detection of B7H4 and HER2 expression in gastric cancer cells. (A) Protein expression of B7H4 and HER2 in cells. (B,C) Detection of B7H4 mRNA and protein expression following B7H4 knockdown. ****, P<0.0001. GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GES, gastric epithelial cells; HER2, human epidermal growth factor receptor 2; NC, negative control. Assessment of the efficiency of short hairpin RNA (shRNA) in knocking down B7H4 in NCI-N87 Lentiviral transfection was performed using the previously screened NCI-N87 cells to knock down B7H4 expression. Synthesised shRNA was transfected into NCI-N87 cells, and changes in B7H4 expression levels were systematically assessed. Results from qRT-PCR and Western blot experiments both indicated that, compared with sh-B7H4#NC, sh-B7H4#1 and sh-B7H4#2 significantly suppressed the expression of B7H4 mRNA and B7H4 protein in NCI-N87 cells, whilst sh-B7H4#3 exhibited a relatively minor inhibitory effect on B7H4 gene and protein expression (Figure 2B,2C). Effect of B7H4 on the proliferation capacity of HER2-positive GC cells To elucidate the role of B7H4 in the malignant phenotype of HER2-positive GC cells, we performed colony formation assays and CCK-8 cell proliferation assays on NCI-N87 cells following B7H4 knockdown, systematically evaluating the impact of B7H4 knockdown on cell proliferation capacity. The results of the colony formation assay showed that, compared with the negative control group (sh-B7H4#NC), the number of cell colonies in the B7H4 knockdown groups (sh-B7H4#1 and sh-B7H4#2) was significantly reduced; statistical analysis revealed that this difference was statistically significant (as shown in Figure 3A,3B). These results indicate that reducing B7H4 expression effectively inhibits the in vitro clonogenic capacity of NCI-N87 cells, suggesting that B7H4 plays a crucial role in maintaining the long-term proliferation of HER2-positive GC cells. To dynamically monitor the effect of B7H4 knockdown on the short-term proliferative activity of tumor cells, a 72-hour CCK-8 assay was conducted. Cell growth curves were plotted by continuously measuring absorbance values at different time points. The results showed that after 72 hours of culture, the cell proliferation activity in the B7H4 knockdown groups (sh-B7H4#1 and sh-B7H4#2) was significantly lower than that in the control group (sh-B7H4#NC), and statistical analysis confirmed the significance of this difference (Figure 3G). This result corroborates the results of the clonogenic assay, collectively indicating that knocking down B7H4 expression significantly inhibits the proliferation rate of HER2-positive GC cells. Effects of B7H4 knockdown on cell proliferation, migration, and invasion. (A,B) Colony formation assays evaluated the effect of B7H4 knockdown on cell cloning ability. Cells were stained with crystal violet. (C,D) Scratch assays evaluated the effect of B7H4 knockdown on cell migration ability (×100). (E,F) Transwell assays evaluated the effect of B7H4 knockdown on cell migration and invasion abilities. Cells were stained with crystal violet (×200). (G) CCK-8 assay evaluated the effect of B7H4 knockdown on cell proliferation. ***, P<0.001; ****, P<0.0001. CCK-8, Cell Counting Kit-8; NC, negative control; OD, optical density. Effect of B7H4 on the migration and invasion capabilities of HER2-positive GC cells To investigate the effect of B7H4 on the motility, migration and invasion capabilities of HER2-positive GC cells, this study further conducted cell scratch healing assays and Transwell migration/invasion assays. The results of the cell scratch assay showed that, 24 hours after scratching, the scratch gap in the sh-B7H4#NC group healed more rapidly. In contrast, the B7H4 knockdown groups (sh-B7H4#1 and sh-B7H4#2) exhibited a markedly slower wound healing rate, with the width of the wound gap being significantly greater than that of the control group (P<0.001, Figure 3C,3D). These results indicate that knocking down B7H4 effectively inhibits the planar migration capacity of HER2-positive GC cells. The results of the Transwell migration/invasion assays consistently showed that, compared with the negative control group (sh-B7H4#NC), the number of migrating and invading cells in the sh-B7H4#1 and sh-B7H4#2 groups was significantly reduced (P<0.001) (Figure 3E,3F). This indicates that knocking down B7H4 not only impairs cellular motility but also significantly inhibits the ability of tumor cells to actively invade surrounding tissues. Potential mechanisms of action and signalling pathways of B7H4 in HER2-positive GC To elucidate the molecular mechanisms by which B7H4 regulates the malignant biological behaviour of HER2-positive GC cells, this study conducted a systematic investigation based on cellular functional assays, combined with high-throughput sequencing and validation of key signalling pathways. To characterise the changes in the whole-genome expression profile following B7H4 knockdown, we performed transcriptomic sequencing (RNA-seq) on NCI-N87 cells from the sh-B7H4#NC, sh-B7H4#1 and sh-B7H4#2 groups. The sequencing results revealed that, compared with the control group, a total of 557 genes exhibited significant changes following B7H4 knockdown, with 315 genes downregulated and 242 genes upregulated (Figure 4A). These global changes in gene expression preliminarily confirm that B7H4 exerts a broad regulatory role at the transcriptional level. GO functional enrichment analysis of differentially expressed genes revealed that the biological processes significantly affected by B7H4 knockdown were primarily concentrated in cell proliferation, cell migration, cell invasion, and epithelial-mesenchymal transition (EMT) (Figure 4C). This provides bioinformatic validation of our previous functional experimental results and offers molecular-level evidence for the phenotype of B7H4 promoting proliferation, migration, and invasion. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis further revealed potential signalling networks regulated by B7H4. Notably, the nuclear factor κB (NF-κB) signalling pathway was significantly enriched (Figure 4B). The NF-κB pathway is a known key hub regulating cell survival, proliferation, inflammatory responses, EMT and metastasis. These findings suggest that B7H4 may exert its oncogenic function by modulating the NF-κB signalling pathway. Based on GO analysis indicating that the EMT process is significantly affected, we performed Western blot analysis to detect core markers of EMT. The results showed that, compared with the negative control group (sh-B7H4#NC), the expression levels of the mesenchymal phenotype markers N-cadherin and vimentin were significantly reduced in the B7H4 knockdown groups (sh-B7H4#1 and sh-B7H4#2), whilst the expression levels of the epithelial phenotype markers E-cadherin and vimentin were significantly elevated (Figure 4D-4G). This protein-level evidence directly confirms that B7H4 knockdown can reverse the EMT process in HER2-positive GC cells, i.e., it promotes the reversal of cells from a highly migratory and invasive mesenchymal state back to an epithelial state. To validate the predictions of the KEGG analysis, we further examined key indicators of NF-κB pathway activation. Phosphorylation of the NF-κB p65 subunit (p-p65) is a critical step in its activation and nuclear translocation to exert transcriptional regulatory functions. Western blot results showed that, compared with the control group, the expression level of p-p65 protein was significantly reduced in the B7H4 knockdown group, whilst the total p65 protein level remained unchanged (Figure 4H-4J). This indicates that B7H4 promotes the phosphorylation and activation of the NF-κB signalling pathway. Effects of B7H4 knockdown on cellular signalling pathways. (A) A heatmap showing differentially expressed genes following B7H4 knockdown. (B) KEGG enrichment analysis showing the effects of B7H4 knockdown on cellular signalling pathways. (C) GO enrichment analysis showing the effects of B7H4 knockdown on cellular biological functions. (D-G) Western blot analysis of the effects of B7H4 knockdown on the expression of EMT-related markers. (H-J) Western blot analysis of the effects of B7H4 knockdown on the NF-κB signalling pathway. ns, P>0.05; ****, P<0.0001. BP, biological process; CC, cellular component; EMT, epithelial-mesenchymal transition; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; MF, molecular function; NC, negative control; NF-κB, nuclear factor κB. The above findings demonstrate that B7H4 regulates the expression of its downstream target genes by activating the NF-κB signalling pathway. On the one hand, this drives the EMT process (manifested as downregulation of E-cadherin and upregulation of N-cadherin and vimentin), conferring enhanced migratory and invasive capabilities upon the cells; on the other hand, it may simultaneously promote cell proliferation, ultimately contributing to the phenotypic characteristics of tumor malignant progression. Regulation of B7H4 expression influences the therapeutic sensitivity of HER2-positive GC cells to trastuzumab Based on the aforementioned findings regarding the key role of B7H4 in the malignant progression of HER2-positive GC, this study further explored the most critical question regarding its clinical translational value: whether the expression level of B7H4 influences the therapeutic sensitivity of HER2-positive GC to trastuzumab. To this end, we evaluated the impact of B7H4 knockdown on the efficacy of trastuzumab through pharmacological intervention experiments. We first employed a clonogenic assay to simulate conditions of long-term drug exposure, assessing the effect of B7H4 knockdown in combination with different concentrations of trastuzumab (0, 5, 10, 20, 50, 100 µg/mL) on the viability of NCI-N87 cells. The experimental results demonstrated: (I) dose-dependence: in both the sh-B7H4#NC group (control group) and the sh-B7H4#1 group (experimental group), the number of cell colonies formed showed a gradual decline as the concentration of trastuzumab increased, consistent with a concentration-response relationship. (II) Synergistic sensitising effect: at the same concentration of trastuzumab, the colony-forming capacity of the B7H4-knockdown experimental group was significantly lower than that of the control group. This result indicates that reducing B7H4 expression not only inhibits cell proliferation but also significantly enhances the cells’ sensitivity to trastuzumab, producing a synergistic inhibitory effect (Figure 5A,5B). Effect of B7H4 knockdown on the efficacy of trastuzumab. (A,B) Clonogenic assays assessing the effect of B7H4 knockdown on the clonogenic capacity of cells in response to different concentrations of trastuzumab. Cells were stained with crystal violet. (C) CCK-8 assay demonstrating the effect of different concentrations of trastuzumab on cell proliferation. ***, P<0.001; ****, P<0.0001. CCK-8, Cell Counting Kit-8; IC50, half-maximal inhibitory concentration; NC, negative control. To precisely quantify the extent of the effect of B7H4 knockdown on sensitivity to trastuzumab, the study further assessed the short-term drug response of the cells via a CCK-8 cell viability assay and calculated the half-maximal inhibitory concentration (IC50), a key pharmacodynamic parameter. The results showed that the IC50 value for cells in the negative control group (sh-B7H4#NC) was 76.68 µg/mL, whereas the IC50 value for cells in the B7H4 knockdown group (sh-B7H4#1) was significantly reduced to 17.35 µg/mL. This approximately 4.4-fold reduction in the IC50 value provides strong quantitative evidence that B7H4 knockdown significantly enhances the sensitivity of HER2-positive GC cells to trastuzumab (Figure 5C). The above experimental results indicate that knocking down B7H4 significantly enhances the cytotoxic effect of trastuzumab on HER2-positive GC cells and reduces the half-maximal inhibitory concentration by approximately 4.4-fold, which has clear positive implications for improving the sensitivity of HER2-positive GC to targeted therapy. Discussion GC ranks among the leading causes of cancer incidence and mortality worldwide; its high biological heterogeneity and poor prognosis present significant clinical challenges (11,12). As a distinct subtype of GC, HER2-positive GC has seen improved patient survival with the use of the first-line targeted drug trastuzumab; however, unexplained drug resistance remains a key factor limiting further improvements in efficacy (13). Consequently, there is an urgent need for research into new targets that can synergise with or sensitise HER2 therapy. The immune checkpoint molecule B7H4 has been found to be abnormally expressed in various tumors and exhibits direct oncogenic functions independent of immune regulation (14). However, the biological function and mechanism of action of B7H4 in HER2-positive GC remain unclear. To investigate the potential role of B7H4 in HER2-positive GC, this study first analysed TCGA data and found that B7H4 and HER2 mRNA expression in GC tissues were significantly positively correlated. Subsequent validation in 50 clinical samples of HER2-positive GC revealed that approximately 48% of samples co-expressed B7H4 protein, with protein levels also showing a significant positive correlation. Further analysis revealed that high B7H4 expression is associated with pathological features indicative of high invasiveness and poor prognosis, suggesting that B7H4 may promote the proliferation, invasion and metastasis of HER2-positive GC cells. Subsequently, the study demonstrated the oncogenic role of B7H4 through cellular functional assays. Results from colony formation assays and CCK-8 proliferation assays clearly demonstrated that B7H4 knockdown significantly inhibited both the long-term proliferative capacity and short-term proliferation rate of NCI-N87 cells. This not only confirmed that B7H4 is essential for maintaining the proliferation of this cell line but also suggested that it may play a key role in the early stages of GC development or during sustained growth. More importantly, the results of scratch assays and Transwell migration/invasion assays collectively reveal that B7H4 drives the migration and invasion capabilities of GC cells, forming the cellular biological basis for local tumor infiltration and distant metastasis. This indicates that B7H4 not only promotes the proliferation of HER2-positive GC cells but is also a key driver in their acquisition of a highly invasive phenotype. Complex molecular networks inevitably underlie cellular functional phenotypes. To elucidate the mechanism by which B7H4 exerts its effects, this study employed RNA-seq technology. The sequencing analysis revealed that the GO analysis of differentially expressed genes was enriched in processes such as cell proliferation, migration, invasion and EMT. This provides bioinformatic validation of the cellular phenotypes observed in our functional experiments, thereby establishing a closed-loop relationship between functional and molecular changes. The results of the KEGG pathway enrichment analysis clearly pointed to the NF-κB signalling pathway. NF-κB is a classic family of transcription factors that plays a central role in regulating inflammation, immune responses, cell survival, proliferation, and EMT (14,15). In tumors, the sustained activation of NF-κB is closely associated with tumor initiation, progression, metastasis, and resistance to treatment (13,16). Western blot validation experiments in this study confirmed that knockdown of B7H4 significantly reduced the phosphorylation level of the NF-κB p65 subunit (p-p65) without affecting its total protein levels. This directly demonstrates that B7H4 positively regulates the activation of the NF-κB signalling pathway. One of the classic downstream effects of the NF-κB pathway is the promotion of EMT (17,18). EMT is a key biological process whereby tumor cells of epithelial origin acquire mesenchymal characteristics, thereby enhancing their ability to migrate, invade and resist apoptosis. At the molecular marker level, this is manifested by the downregulation of E-cadherin and the upregulation of N-cadherin and vimentin (19,20). The results of this study show that following B7H4 knockdown, E-cadherin expression significantly rebounded, whilst N-cadherin and vimentin expression significantly decreased. These findings indicate that B7H4 acts as an upstream switch driving EMT in HER2-positive GC cells, and that B7H4 knockdown can effectively reverse the EMT process. The mechanisms underlying trastuzumab resistance are complex and involve compensatory activation of the HER2 signalling pathway (such as activation of other HER family members) (21), abnormalities in downstream signalling pathways (such as PI3K/AKT/mTOR) (22), phenotypic conversion (23), and alterations in the tumor microenvironment (24,25). This study found that B7H4 plays a key role in regulating sensitivity to trastuzumab treatment. Drug sensitivity assay results indicated that, under treatment with the same concentration of trastuzumab, cell viability in the B7H4-knockdown group was significantly lower than in the control group, suggesting that B7H4 knockdown produces a synergistic cytotoxic effect with trastuzumab. Analysis of IC50 values via the CCK-8 assay revealed that B7H4 knockdown reduced the IC₅₀ value of NCI-N87 cells to trastuzumab by approximately 4.4-fold (from 76.68 to 17.35 µg/mL). This suggests that high B7H4 expression is one of the key factors contributing to the reduced sensitivity of HER2-positive GC cells to trastuzumab, and that inhibition of B7H4 can effectively reverse or partially reverse this resistance. Moreover, although in vitro data strongly suggest that B7H4 knockdown exerts a sensitizing effect in vivo, independent validation using patient-derived xenograft (PDX) models or subcutaneous xenograft models remains essential to substantiate this conclusion. The aforementioned mechanisms by which B7H4 influences trastuzumab sensitivity may be associated with EMT, activation of the NF-κB pathway, and the B7H4-mediated inhibitory tumor microenvironment. Numerous studies have confirmed that EMT is closely associated with resistance to various targeted therapies and chemotherapy; cells undergoing EMT exhibit stronger stem cell-like properties, anti-apoptotic capacity, and drug efflux capabilities (26-28). In this study, B7H4-driven EMT may represent a key mechanism underlying the insufficient sensitivity to trastuzumab. Knocking down B7H4 reverses EMT, restoring cells to an epithelial state, which may consequently enhance their sensitivity to targeted therapies. Furthermore, studies have shown that activation of the NF-κB pathway interferes with HER2-targeted therapy by upregulating anti-apoptotic proteins and promoting cell cycle progression (29-32). By activating NF-κB, B7H4 may establish a network of mechanisms that inhibit apoptosis and confer drug resistance. Current evidence indicates that a direct mechanistic link between B7-H4 and trastuzumab resistance remains insufficiently established. Previous studies suggest that B7-H4 may intersect with canonical mechanisms of trastuzumab resistance, including PI3K/AKT signaling, PTEN loss, and the interleukin (IL)-6/STAT3 axis. The PI3K/AKT/mTOR pathway represents a central downstream signaling cascade implicated in trastuzumab resistance. Loss of PTEN can relieve its inhibitory control over PI3K/AKT signaling and has been associated with reduced clinical efficacy of trastuzumab (33). Studies have shown that B7-H4 expression is closely associated with an epithelial cellular state and is regulated by PI3K activity. In animal models, cell-surface B7-H4 expression enhances tumor resistance to anti-PD-L1 therapy and is associated with an immune-cold tumor microenvironment (34). These findings provide a plausible conceptual bridge between B7-H4 and trastuzumab resistance: sustained activation of PI3K/AKT signaling may not only maintain tumor cell survival and proliferation, but may also upregulate or stabilize immunosuppressive molecules, thereby promoting immune evasion. On the other hand, the relationship between B7-H4 and tumor stemness, EMT, and dedifferentiation appears to be complex. One study demonstrated that SOX9 can activate STAT3 and directly induce B7-H4 expression, thereby enabling dedifferentiated tumor cells to evade immune surveillance. In human breast cancer, SOX9 expression was correlated with B7-H4 expression and was accompanied by reduced CD8+ T-cell infiltration (35). In addition, B7-H4 functions as a co-inhibitory immune checkpoint molecule that suppresses T-cell proliferation and cytokine secretion. This effect involves inhibition of downstream TCR/CD28 signaling pathways, including ERK, JNK, p38, and AKT (36). In tumor models, blockade of B7-H4 has been shown to restore tumor antigen-specific T-cell activation and delay tumor growth (37). These findings suggest that high expression of B7-H4 by tumor cells or tumor-associated macrophages may impair effector T-cell activation and limit cytotoxic immune responses, thereby weakening the immune-mediated tumor clearance upon which trastuzumab therapy partly depends. In other words, B7-H4 may not directly interfere with the binding of trastuzumab to HER2; rather, it may attenuate the immune effector functions of HER2-targeted antibodies by shaping an immunosuppressive tumor microenvironment. Moreover, IL-6 and IL-10 have been reported to induce B7-H4 expression in macrophages (38), suggesting that B7-H4 may not necessarily function downstream of PI3K/AKT activation or PTEN loss, but may instead act as an effector molecule within an inflammation-driven immunosuppressive network, amplifying microenvironmental alterations triggered by tumor cell-intrinsic resistance mechanisms. Studies have further shown that B7-H4 can promote PD-L1 transcription by modulating DNMT1-mediated methylation of the PD-L1 promoter and participate in interferon (IFN)-γ-induced PD-L1 expression (39). B7-H4 has also been implicated in immune evasion and in attenuating the therapeutic efficacy of PD-1/PD-L1 blockade (40,41). A cohort study of 110 patients with HER2-positive GC (all receiving chemotherapy plus trastuzumab) reported that high expression of ERBB2d16, a splice variant of HER2, was associated with several immunosuppressive features, including elevated PD‑L1 expression (P=0.0003), reduced CD3+ T-cell infiltration (P=0.04), decreased T-cell receptor clonal diversity (P=0.04), and aberrant EMT markers (42). These mechanistic observations raise the possibility of an overlap with the B7H4-induced immunosuppressive microenvironment, suggesting that B7H4 and ERBB2d16 might act synergistically to promote resistance to anti-HER2 therapy—although it should be noted that B7H4 expression was not assessed in that study. Taken together, these findings appear to indicate that B7H4 does not operate in isolation; rather, it may cooperate with a broader immunosuppressive network to constitute a therapeutic-resistant niche. In the context of trastuzumab treatment, such immune-desert or immune-excluded states could potentially restrict the infiltration of NK cells, CD8+ T cells, and other effectors into the tumor, thereby possibly attenuating the in vivo efficacy of anti-HER2 antibodies. In summary, the currently available evidence does not yet permit a definitive conclusion that B7H4 is a direct driver of trastuzumab resistance. The relevant inferences are largely derived from studies investigating B7H4 in the contexts of tumor immune evasion, regulation of drug-resistance signaling, tumor stemness, and resistance to immunotherapy. The therapeutic activity of trastuzumab relies on both HER2 signal blockade and antibody-dependent cellular cytotoxicity (ADCC), which necessitates the coordinated participation of NK cells and T cells (43). High B7H4 expression could theoretically attenuate ADCC activity—and thereby reduce drug responsiveness—by suppressing T-cell function and promoting Foxp3+ regulatory Treg-cell infiltration; however, this putative causal chain remains to be directly validated in the setting of GC. Moreover, the EMT and PD‑L1 upregulation induced by the HER2 splice variant ERBB2d16 may potentially synergize with the “cold” tumor microenvironment shaped by B7H4, and similar immune-excluded patterns have been associated with resistance to immune checkpoint inhibitors in breast cancer, although whether this scenario applies to GC has yet to be determined (42). It is also noteworthy that, in GC, HER2-positive regions and PD-L1-expressing regions are spatially distinct, which raises the possibility that B7-H4 may likewise exhibit substantial intratumoral heterogeneity; consequently, a single biopsy may not accurately reflect its overall expression status, posing a challenge for biomarker evaluation. Currently, interventional strategies targeting B7H4—including monoclonal antibodies, antibody-drug conjugates, and CAR‑T cell therapies—are still in the exploratory stage. Should subsequent systematic analyses be performed in trastuzumab-sensitive and -resistant models, incorporating assessments of PI3K/AKT/PTEN/MET status, HER2 expression levels, ADCC sensitivity, tumor-infiltrating lymphocyte (TIL)/NK cell infiltration, and immunosuppressive factors such as PD‑L1, and should these analyses help clarify whether B7H4 functions as a resistance driver, a companion biomarker, or a node of immune evasion, then a combination regimen of trastuzumab with anti-B7H4 agents might emerge as a promising strategy to overcome resistance or enhance therapeutic efficacy. Nevertheless, it must be acknowledged that the present study is constrained by its reliance on a single cell line for validation and by the depth of mechanistic exploration; accordingly, the conclusions drawn should be considered preliminary. Future independent investigations will be required to extend the validation to multiple cell lines and to incorporate matched negative control cell lines, in order to further refine the experimental system and provide more robust data to support the above hypotheses. Ultimately, the clinical translation of this combination strategy would still depend on more solid mechanistic insights and preclinical evidence. Conclusions In this study, we demonstrated that B7H4 is highly expressed in HER2-positive GC tissues and is significantly coexpressed with HER2. Its expression level positively correlates with adverse clinicopathological features, including larger tumor size, poorer differentiation, deeper invasion, lymph node and distant metastasis, as well as vascular and neural invasion, indicating that B7H4 contributes to malignant progression and poor prognosis in HER2-positive GC. Mechanistically, B7H4 activates the NF-κB signaling pathway and drives EMT, thereby significantly promoting tumor cell proliferation, migration, and invasion. In terms of therapeutic translation, we found that B7H4 expression directly modulates trastuzumab sensitivity: knockdown of B7H4 reduced the IC50 from 76.68 to 17.35 µg/mL, representing an approximately 4.4-fold increase in sensitivity, suggesting that B7H4 is a key mediator of trastuzumab resistance. Collectively, B7H4 not only serves as a potential prognostic biomarker but also represents a novel therapeutic target to overcome trastuzumab resistance in HER2-positive GC. Combined blockade of B7H4 and HER2 signaling may offer a promising strategy to improve clinical outcomes, although further validation in multiple cell lines, in vivo models, and clinical trials is warranted before clinical translation. Supplementary The article’s supplementary files as Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2026-1121/prf