TE-1, −4, −5, −6, −8, −9, −10, −11, −14 and −15 human esophageal squamous cancer cell lines were purchased from the Cell Resource Center for Biomedical Research, Institute of Development, Aging and Cancer, Tohoku University (Sendai, Japan). All cells were cultured in RPMI-1640 medium (Thermo Fisher Scientific, Inc.) containing 10% (v/v) FBS (Thermo Fisher Scientific, Inc.) and 1% (v/v) penicillin-streptomycin solution (Thermo Fisher Scientific, Inc.) at 37°C in a humidified incubator with 5% CO₂. To evaluate morphological differences among these cell lines, cells at 50–70% confluency were fixed and stained using a Diff-Quick staining kit (Sysmex Corporation). The cells were sequentially immersed for 2 min each in the fixative solution, stain solution I and stain solution II at room temperature. Stained cells were observed under a BZ-9000 light microscope (Keyence Corporation) at a magnification of ×20 and analyzed using the dedicated BZ-H3C software (version 1.4.1; Keyence Corporation).

Peripheral venous blood samples were collected from healthy adult donors (male; median age, 38 years; age range, 36–60 years) at Hokkaido University (Sapporo, Japan) between July 2024 and September 2024. All participants provided written informed consent prior to participation. Inclusion criteria included age ≥20 years and the ability to provide informed consent. Individuals with a history of cancer or active malignant disease were excluded. The present study was approved (approval no. 24-0126; approval date July 10, 2024) by the Institutional Review Board Committee of Hokkaido University (Sapporo, Japan). Blood collection from healthy donors began in July 2024, and the samples were subsequently used for experiments. PBMCs were isolated from heparinized blood samples via density gradient centrifugation at 400 × g for 30 min at room temperature using Ficoll-Paque Plus reagent (GE Healthcare) according to the manufacturer's instructions.

Cells were lysed using an ULTRARIPA kit (BioDynamics Laboratory, Inc.) supplemented with a Protease Inhibitor Cocktail (Promega Corporation). Protein concentrations were determined using the TaKaRa BCA Protein Assay Kit (Takara Bio Inc.). Equal amounts of protein (10 µg per lane) were loaded onto 15% SDS-PAGE gels and transferred to PVDF membranes (Bio-Rad Laboratories, Inc.). Human PBMCs were used as positive controls. After protein transfer to the PVDF membranes, the membranes were blocked with Tris-buffered saline with 0.1% Tween-20 (TBST) and 5% skim milk at room temperature for 1 h, and then incubated with primary antibodies overnight at 4°C. A list of the antibodies used in the present study (including dilutions, cat. nos. and suppliers) is presented in Table I. After two washes with TBST, the membranes were incubated at room temperature for 1 h with HRP-conjugated secondary antibodies, including anti-rabbit IgG (HRP-linked; cat. no. 7074; Cell Signaling Technology, Inc.) and anti-mouse IgG (HRP-linked; cat. no. 7076; Cell Signaling Technology, Inc.) antibodies, depending on the host species of the primary antibody, both diluted 1:4,000 in TBST containing 5% skim milk. Detection was performed using an ECL plus enhanced chemiluminescent substrate (GE Healthcare) and a ChemiDoc instrument (Bio-Rad Laboratories, Inc.). Densitometry was performed using Image Lab software (version 6.1.0; Bio-Rad Laboratories, Inc.). Equal loading was confirmed using actin staining.

Total RNA was extracted from TE cell lines or human PBMCs using the RNeasy Mini Kit (Qiagen GmbH) according to the manufacturer's protocol as previously described (22). cDNA was generated using 1 µg RNA per reaction with Prime Script RT Master Mix (Takara Bio, Inc.) according to the manufacturer's protocol. RT-qPCR was performed using TaqMan Universal PCR Master Mix (No AmpErase UNG; Thermo Fisher Scientific, Inc.) and TaqMan probes (human CD40, Hs01002915_g1; human MMP-9, Hs00957562_m1; and human β-actin, Hs99999903_m1) on a Lightcycler instrument (Roche Diagnostics). The sequences of the primers and probes used in these assays are proprietary and not publicly disclosed by the manufacturer. The thermocycling conditions were as follows: Initial enzyme activation at 95°C for 10 min, followed by 40 cycles of denaturation at 95°C for 15 sec and annealing/extension at 60°C for 1 min. Data were analyzed using the 2^−ΔΔCq method (23), and β-actin was used as the reference gene for normalization.

CD40 mRNA expression was evaluated using semi-quantitative PCR. Total RNA was extracted from TE cells using the RNeasy Mini Kit (Qiagen GmbH) and quantified spectrophotometrically. Reverse transcription was performed using 1 µg RNA with the Prime Script RT Master Mix (Takara Bio, Inc.) according to the manufacturer's protocol. The reverse transcription conditions were 37°C for 15 min and 85°C for 5 sec. The PCR reaction was performed with KOD Plus polymerase (Toyobo Co., Ltd.), which includes dNTPs in the reaction mixture, according to the manufacturer's instructions. The thermocycling conditions were as follows: 94°C for 2 min; 35 cycles of 98°C for 10 sec, 64°C for 1 sec and 68°C for 28 sec; followed by a hold at 10°C. The following primers were used: CD40 forward, 5′-AGAGTTCACTGAAACGGAATGCC-3′ and reverse, 5′-ACAGGATCCCGAAGATGATGG-3′; and β-actin forward, 5′-CAACCGCGAGAAGATGACCC-3′ and reverse, 5′-GGAACCGCTCATTGCCAATGG-3′. PCR products were separated on 2% agarose gels and visualized using ethidium bromide staining under UV light with a ChemiDoc instrument (Bio-Rad Laboratories, Inc.). Densitometric analysis was performed using Image Lab software (version 6.1.0; Bio-Rad Laboratories, Inc.). β-actin was used as the reference gene to confirm equal loading and for normalization.

The Prime PCR 96-well plate array ECM Remodeling H96 (cat. no. 10025275; Bio-Rad Laboratories, Inc.) was used according to the manufacturer's instructions to evaluate ECM-related gene expression. PCR was performed using the Lightcycler instrument (Roche Diagnostics) as aforementioned, and data were analyzed using the ΔΔCq method. TE-5 and TE-10 EC cell lines were used in this assay.

The cells were first stained for the surface antigens CD40, CD154, CD61 and CD62P, followed by washing with autoMACS Rinsing Solution (Miltenyi Biotec GmbH). Cells were surface-stained for 10 min at 4°C with the following antibodies: CD40-VioBright™ FITC (cat. no. REA733; 1:50), CD154-allophycocyanin (cat. no. REA238; 1:50), CD61-VioBright™ 515 (cat. no. REA761; 1:50) and CD62P-phycoerythrin (cat. no. REA389; 1:500) (all from Miltenyi Biotec GmbH). Data were acquired using a MACSQuant^® Analyzer 10 flow cytometer (Miltenyi Biotec GmbH) and analyzed using FlowJo software (version 10.6.2; Becton, Dickinson and Company).

Cells were cultured on coverslips and fixed with 3.7% formaldehyde for 15 min at 25°C. After three washes with PBS, the cells were blocked with PBS containing 3% BSA (Thermo Fisher Scientific, Inc.) and 10% (v/v) goat serum (Thermo Fisher Scientific, Inc.) for 15 min at 25°C. The cells were then incubated with human polyclonal anti-CD40 antibodies (1:500; cat. no. ab13545; Abcam) in PBS for 1 h at 25°C. After three PBS washes, the cells were incubated with Alexa Fluor^® 488-conjugated goat anti-rabbit antibodies (1:500; cat. no. ab150077; Abcam) for 1 h at 25°C. After three PBS washes, the coverslips were mounted on slides using Dapi-Fluoromount-G™ (Southern Biotech), and the nuclei were labeled by staining with 4′,6-diamidino-2-phenylindole. Permeabilization was not performed, as CD40 is a membrane protein with an extracellular epitope recognized by the antibody used. DAPI counterstaining was performed according to the manufacturer's instructions, which included incubation at room temperature for 5 min. Finally, the samples were analyzed using a confocal microscope (BZ-9000; Keyence Corporation) with the analysis software BZ-H3C (version 1.4.1; Keyence Corporation).

Cells were treated with 5 µg/ml recombinant soluble CD154 (rsCD154; ENZO MEGACD40L; Enzo Life Sciences) for 24 h in serum-free culture medium. After incubation at 37°C, the supernatants were harvested and centrifuged at 10,000 × g for 5 min at 4°C. Finally, the cell pellets were lysed using RNeasy kit (Qiagen GmbH), as described in the RT-qPCR section, for total RNA extraction. For the dose-response assay, TE-10 cells were treated under the same conditions with various concentrations of rsCD154 (10 ng/ml, 30 ng/ml, 100 ng/ml, 300 ng/ml, 1 µg/ml, 3 µg/ml, 10 µg/ml and 30 µg/ml) to evaluate MMP-9 mRNA expression in response to CD154 stimulation. The expression levels of MMP-9 mRNA were quantified using RT-qPCR as aforementioned.

The CellTiter 96 AQueous One Solution Cell Proliferation Assay System (Promega Corporation) was used according to the manufacturer's instructions for the cell viability assays. Briefly, 4×10³ cells/well were plated into a 96-well plate and incubated for 48 h with or without stimulation with 5 µg/ml rsCD154 at 37°C in a humidified incubator with 5% CO₂. After 48 h of incubation, 10 µl CellTiter 96 AQueous One Solution reagent was added to each well. The absorbance was measured at 490 nm using a SpectraMax I3 microplate reader (Molecular Devices, LLC) after 2 h of incubation at 37°C in a humidified 5% CO₂ atmosphere. A total of five replicate wells were used to measure cell viability.

Transwell cell migration assays were performed using 8-µm pore size membrane chambers without Matrigel (Corning BioCoat Control Inserts; cat. no. 354578; Corning, Inc.), while invasion assays were performed using Matrigel-coated chambers of the same type (Corning BioCoat Matrigel Invasion Chambers; cat. no. 354480; Corning, Inc.). A total of 5×10⁴ cells in 200 µl RPMI-1640 medium containing 1% FBS and 25 pg rsCD154 were seeded in the upper compartments, and 500 µl RPMI-1640 medium containing 10% FBS as a chemoattractant was added to the lower compartments. After 48 h of incubation at 37°C in a humidified incubator with 5% CO₂, the non-invading cells were removed from the upper surface of the membrane by scrubbing with cotton-tipped swabs. Invasive cells on the membranes beneath the insert were fixed and stained using a Diff-Quick staining kit (Sysmex Corporation), which consists of a fixative solution, stain solution I and stain solution II. The membranes were sequentially immersed in each solution for 2 min at room temperature, according to the manufacturer's instructions. Stained cells were counted using a confocal microscope (BZ-9000; Keyence Corporation) and KEYENCE software (BZ H3C; version 1.4.1). A total of five fields were randomly selected, the total area of cells in each field was measured and the average was obtained to measure the number of invasive cells.

Culture media from esophageal squamous cell carcinoma (ESCC) cells stimulated with or without 5 µg/ml rsCD154 for 24 h at 37°C were analyzed for gelatin degradation using 7.5% SDS-PAGE gels containing 0.1% (w/v) gelatin (from porcine skin type A; MilliporeSigma). After SDS-PAGE, the gels were washed twice for 1 h in 2.5% Triton X-100 and incubated for 24 h at 37°C in an incubation buffer [50 mM Tris (pH 7.4), 1 µM ZnCl₂, 5 mM CaCl₂ and 1% Triton X-100]. The gels were stained with 0.5% Coomassie brilliant blue R250 in methanol for 1 h at room temperature, and destained with a solution containing 40% methanol and 10% acetic acid for 4 h at room temperature. Gelatinolytic activity appeared as white bands on a blue background. The gels were scanned, and the protein band intensities were determined using a ChemiDoc image analyzer (Bio-Rad Laboratories, Inc.) and Image Lab software (version 6.1.0; Bio-Rad Laboratories, Inc.).

CD40-specific and control siRNA were obtained from Santa Cruz Biotechnology, Inc. Control siRNA-A (cat. no. sc-37007) served as a non-targeting scrambled siRNA. CD40 siRNA (cat. no. sc-29250) was used to specifically knock down CD40 gene expression. The sequences of siRNAs were not provided by the supplier. TE-10 cells in the exponential growth phase were seeded into 96-well plates at a density of 7.5×10³ cells per well, cultured for 24 h, and then transfected with 800 nM siRNA using oligofectamine (Thermo Fisher Scientific, Inc.) and OPTI-MEM I reduced-serum medium (Thermo Fisher Scientific, Inc.) at 37°C for 4 h, according to the manufacturer's instructions. The siRNA concentration was determined based on prior dose-response experiments. After transfection, the cells were further incubated in complete growth medium at 37°C. Transfection efficiency was assessed 72 h post-transfection via flow cytometry using allophycocyanin-conjugated anti-CD40 antibodies for staining. Subsequent experiments were conducted 72 h after transfection. Control cells underwent mock transfection with control siRNA under identical conditions.

TE-10 cells (1×10⁵ cells/well) were seeded in 12-well plates, incubated at 37°C for 24 h and subsequently stimulated with 5 µg/ml rsCD154 for 15 or 30 min at room temperature. Cell pellets were collected at 15 and 30 min after stimulation. Proteins were then extracted from the cell pellets following the aforementioned western blotting protocol. The extracted proteins were analyzed using the antibodies listed in Table I to evaluate the activation of each signaling pathway involved in the CD40-CD154 interaction.

Venous blood was collected in an aplastic syringe containing 1/10 volume of CPD buffer (16 mM citric acid, 90 mM sodium citrate, 16 mM NaH₂PO₄, 142 mM dextrose, pH 7.4). Blood samples were centrifuged at 200 × g for 20 min at 20°C. The top layer containing platelet-rich plasma was transferred into a new tube containing 1 volume of HEP buffer (140 mM NaCl, 2.7 mM KCl, 3.8 mM HEPES, 5 mM EGTA, pH 7.4) and prostaglandin E₁ (1 µM final concentration; Cayman Chemical Company), and centrifuged at 100 × g for 20 min at 20°C. The supernatant was collected and placed into a new tube, which was centrifuged at 800 × g for 20 min at 20°C. After discarding the supernatant, the platelet pellet was washed twice with a platelet wash buffer [10 mM sodium citrate, 150 mM NaCl, 1 mM EDTA and 1% (w/v) dextrose, pH 7.4]. The platelet pellet was slowly resuspended in modified Tyrode buffer (134 mM NaCl, 12 mM NaHCO₃, 2.9 mM KCl, 0.34 mM Na₂HPO₄, 1 mM MgCl₂, 10 mM HEPES, 5 mM dextrose, 3 mg/ml BSA and 1 µM PGE₁, pH 7.4) and used as a platelet solution. Platelets were activated by pre-warming at 37°C for 5 min, followed by the addition of 1 U human α-thrombin (Prolytix). After thrombin addition, the mixture was allowed to stand at room temperature for 1 min.

The concentrations of soluble CD154 (sCD154) and MMP-9 in the culture supernatant were measured using ELISA kits (Quantikine^® ELISA: Human CD40 Ligand/TNFSF5 Immunoassay; cat. no. DCDL40; R&D Systems, Inc.; Quantikine^® ELISA: Human MMP-9; cat. no. DMP900; R&D Systems, Inc.) according to the respective manufacturer's instructions. The optical density at 540 nm was determined using a SpectraMax I3 microplate reader (Molecular Devices, LLC). All samples were analyzed in quadruplicate.

Briefly, 7.5×10⁴ TE-10 cells were inoculated into 24-well plates and incubated at 37°C for 24 h. After PBS washing, RPMI-1640 medium without FBS was added to the cells, and 1×10⁸ of the activated platelets suspended in 0.2 ml modified Tyrode solution were added to 0.4-µm-pore cell culture inserts and incubated at 37°C for another 24 h. After incubation, the culture supernatant and cell pellet were collected, and MMP-9 mRNA levels were determined using RT-qPCR as aforementioned. TaqMan probe-based assays were used (MMP-9: Hs00957562_m1; β-actin: Hs99999903_m1; Thermo Fisher Scientific, Inc.), and primer sequences were not disclosed by the supplier.

Approval was obtained from the Institutional Review Board of Hokkaido University Hospital (approval no. 24-0126; approval date July 10, 2024; Sapporo, Japan). Written informed consent was obtained from all patients who participated in this clinical research. The present study was conducted between August 2014 and October 2017 on 49 consecutive patients who underwent curative esophagectomy for EC at the Department of Gastroenterological Surgery, Hokkaido University Hospital (Sapporo, Japan). Eligible patients were aged ≥20 years and had provided informed consent to store blood samples for research purposes. Patients judged by the investigators to be inappropriate for inclusion were excluded. The survival analysis included 41 men and 8 women. The median age was 67 years, with an age range of 52–85 years. Patient sera were collected either immediately before surgery on the day of the operation or on the day prior to surgery, and stored at −80°C until further analyses. The samples were analyzed in August 2024. Serum MMP-9 and CD154 concentrations were measured using ELISA kits (Quantikine ELISA Human CD40 Ligand, Human MMP-9; R&D Systems, Inc.). Cancer recurrence and survival were analyzed over a minimum follow-up period of 5 years after esophagectomy. Information on cancer recurrence, survival time, TNM classification (Union for International Cancer Control 8th edition) (24) and histological type was extracted from the medical records.

All statistical analyses were performed using GraphPad Prism 9 software (Dotmatics). Pearson's χ² test, the unpaired t-test or the Mann-Whitney U test were used as appropriate to compare different groups. For comparisons involving three groups, one-way ANOVA was used, followed by Bonferroni post hoc analysis when statistical significance was detected. For survival analysis, the log-rank (Mantel-Cox) test was used to compare Kaplan-Meier survival curves. Data are presented as mean ± SD, unless otherwise indicated. All experiments were performed in triplicate (n=3) unless stated otherwise. P<0.05 was considered to indicate a statistically significant difference.

CD40 mRNA expression was analyzed in 10 TE cell lines, among which TE-1, −5, −10 and −15 cells exhibited relatively high expression levels (Fig. 1A). Western blotting was performed to evaluate CD40 protein levels in TE cells. TE-1, −5, −10 and −15 cells showed clear CD40 bands (Fig. 1B). Flow cytometry and immunocytochemistry analyses were also performed to examine the cell surface expression of CD40 on TE cells, which revealed similar results to those obtained by western blotting, except for TE-1 cells, in which the immunocytochemistry result differed (Fig. 1C and D). No morphological differences were observed among the 10 types of TE cells based on CD40 expression, as assessed by Diff-Quick staining and light microscopy (Fig. S1). Due to discrepancies in CD40 expression levels across the different detection methods, such as weak or absent surface staining by immunocytochemistry in TE-1 cells despite positive results by RT-qPCR and western blotting, semi-quantitative PCR was also performed to further validate mRNA expression. Human CD40 is known to have splicing variants resulting from alternative splicing (25). The presence of splicing variants was confirmed in TE cells positive for CD40 mRNA expression (Fig. 1E). Based on these results, particular emphasis was placed on the flow cytometry and immunocytochemistry data. TE-5 and TE-10 cells, which exhibited high CD40 expression intensities, were selected as CD40-overexpressing cells for subsequent experiments, while TE-4 and TE-11 cells, which had low CD40 expression intensities, were used as low CD40 expression cells.

CD40 activation was performed in vitro to assess changes in TE cell activity and investigate the function of CD40 in EC. In the present study, CD154, a CD40 ligand, was used to activate CD40. The four types of TE cells showed no morphological differences (Fig. S1); however, their baseline viability, migration and invasion varied. Notably, TE-11 cells exhibited high viability and invasion (Figs. 2 and S2). TE cells did not exhibit significant changes in proliferative ability, regardless of the presence or absence of CD40 expression (Fig. 2A). No changes in cell migration or invasion were observed in low CD40 expression cells upon stimulation with rsCD154. By contrast, high CD40 expression cells exhibited significantly increased migration and invasion (Figs. 2B and C, and S2). Although baseline differences in viability, migration and invasion were observed among the cell lines investigated in the present study, the results suggested that the CD40-CD154 interaction in TE cells enhanced cell motility without changing their proliferative capacity. These interactions were only observed in high CD40 expression cells.

Changes in gene expression levels in TE-10 cells due to the CD40-CD154 interaction were examined. The genes involved in ECM remodeling were investigated because no change in cell viability was observed, whereas cell migration and invasion were increased. The expression levels of 40 genes related to changes in ECM remodeling were screened. Specifically, following stimulation with rsCD154, the mRNA expression level of MMP-9 increased by 17.55-fold, that of plasminogen (PLG) by 6.44-fold, and that of laminin subunit γ-2 (LAMC2) by 4.95-fold in TE-10 cells (Fig. 3). The analysis was also performed in TE-5 cells, where a similarly strong upregulation of MMP-9 mRNA was observed (Fig. S3). Therefore, MMP-9 expression was investigated in subsequent experiments.

It was determined whether the CD40-CD154 interaction induced MMP-9 secretion. rsCD154 and PBS (control) were added to TE cells, and MMP-9 secretion in the culture supernatant was analyzed via gelatin zymography (Fig. 4A) and ELISA (Fig. 4B). MMP-9 upregulation was observed in the culture supernatants of high CD40 expression cells, whereas MMP-9 expression in low CD40 expression cells did not change significantly. TE-10 and TE-11 cells, which exhibited higher baseline invasive abilities in migration and invasion assays, also exhibited elevated MMP-9 levels even without sCD154 stimulation. Next, MMP-9 regulation was assessed at the mRNA level. After rsCD154 stimulation, MMP-9 mRNA expression was upregulated in high CD40 expression TE cells, whereas no change was observed in low CD40 expression cells (Fig. 4C). Additionally, the upregulation of MMP-9 mRNA induced by CD154 stimulation occurred in a dose-dependent manner (Fig. S4).

To elucidate the role of the CD40 gene in TE cells, CD40-knockdown TE-10 cells were generated. Transfection of TE-10 cells with si-CD40 or si-control resulted in the creation of CD40 knockdown TE-10 cells and control TE-10 cells (Mock), respectively. The knockdown efficiency was then evaluated by measuring CD40 mRNA levels using qPCR (Fig. 5A). Additionally, the response of these cells to sCD154 stimulation was assessed by measuring MMP-9 mRNA levels. CD40-knockdown TE-10 cells consistently exhibited low MMP-9 mRNA levels regardless of sCD154 stimulation (Fig. 5B). Furthermore, migration and invasion assays were performed using CD40-knockdown TE-10 cells. The results demonstrated that CD40-knockdown TE-10 cells exhibited reduced migratory and invasive capabilities compared with the control cells (Fig. 5C and D).

Given that the CD40-CD154 interaction induces MMP-9 upregulation, the activation of PI3K, MAPK and NF-κB was investigated following CD40 activation in TE-10 cells. As shown in Fig. 6A, phosphorylation of Akt, a downstream effector of PI3K, was observed in CD154-stimulated TE-10 cells, confirming PI3K activation. Furthermore, phosphorylation of Erk1/2 and JNK, indicating MAPK pathway activation, is also shown in Fig. 6A. Nuclear translocation of NF-κB p65 was also observed, as shown in Fig. 6B, indicating its activation.

In vivo, CD154 exists either as a transmembrane protein on the surface of CD4⁺ T lymphocytes or as a soluble form. The majority of circulating sCD154 is derived from platelets (26), which are activated by proteinaceous platelet-activating factors secreted by cancer cells in the tumor microenvironment (TME). The viability and metastasis of cancer cells are promoted by the induction of platelet aggregation (26–28). Therefore, the relationship between platelets and high CD40 expression EC cells was evaluated. Platelet solutions were prepared using blood samples collected from healthy human donors. Because platelets are activated in the TME and engage tumor cells (29), α-thrombin was used to activate platelets in the present study.

The purity of the generated platelet solution was confirmed via flow cytometry for platelet marker CD61 expression (30). Platelets were activated via α-thrombin stimulation, and their activity was confirmed by detecting CD62P (P-selectin) expression, an activated platelet marker (Fig. S5). Thrombin-activated platelets exhibited a slight increase in CD154 expression on their cell surface (Fig. S6). In addition, sCD154 release into the platelet solution was increased (Fig. S7). Next, four types of TE cells were co-cultured with the platelet solution to observe the reaction between the platelet solution and EC cells. MMP-9 mRNA upregulation was confirmed in all four types of TE cells; however, the increase was statistically significant only in TE-5 and TE-10 cells, which exhibited high CD40 expression (Fig. 7).

Esophagectomies were performed on 49 patients during the study period. The median age of the patients was 67 years, and 41 patients were male. The 3-year overall survival (OS) rate was 66.3%, and the 3-year disease-free survival (DFS) rate was 58.5% (Table II). The MMP-9 concentration was 534.1±384.3 ng/ml (mean ± SD), and the CD154 concentration was 2.15±1.12 ng/ml (mean ± SD). In the univariate analysis, platelet counts were significantly higher in the high MMP-9 group, and a trend toward elevated platelet counts was observed in the high CD154 group (Tables SI and SII). The relationship between the survival period and the concentrations of MMP-9 and CD154 was evaluated using Kaplan-Meier curves, with patients stratified into high and low expression groups based on the median value. No significant association was observed between serum MMP-9 or CD154 levels and OS or DFS (Fig. 8A-D). However, when survival was analyzed based on serum MMP-9 and serum CD154 levels stratified by pathological stage (pStage I and II–IV), a significant improvement in prognosis was noted for patients with low MMP-9 levels in pStage I (Fig. 9A and C). By contrast, low MMP-9 levels were significantly associated with reduced OS in patients with pStage II–IV (Fig. 9G). Serum CD154 levels did not exhibit any association with survival, even when analyzed according to pStage.