Abstract
Advanced-stage sarcomas pose a major challenge in oncology, as they are often resistant to conventional therapies and associated with poor prognosis. CAR-based cellular immunotherapy is emerging as a very promising therapeutic option; however, clinically relevant animal models are urgently needed to accelerate the clinical development of these approaches. B7-H3 molecule, due to its low expression in normal tissue, high prevalence in multiple human cancers, and association with cancer stemness and aggressiveness, represents one of the most attractive targets for CAR-immunotherapy. In this study, we established a preclinical cellular immunotherapy platform based on canine cytokine-induced killer cells (CIK) redirected by an antihuman B7-H3 CAR against canine sarcoma cells and 3D sarcoma spheroids. B7-H3 was consistently detected across all analyzed canine sarcoma subtypes, including osteosarcoma, soft tissue sarcoma, and hemangiosarcoma, although with variable levels of expression intensity. We successfully generated canine B7-H3-CAR.CIK, achieving a mean CAR expression of 39% ± 2 with an immune phenotype unmodified (NTD) control (CD3 = 92% ± 3, CD8 = 87% ± 7; CD4 = 49% ± 5; CD5 = 77% ± 0.1; NKp46 = 83% ± 5). Canine B7-H3-CAR.CIK efficiently killed canine sarcoma cell lines compared with NTD.CIK, even at low effector/target (E/T) ratios (B7-H3-CAR.CIK: 45% vs 8%; E:T 1:1; 48 h; N = 7, n = 8; p < 0.0001), and demonstrated significant cytotoxicity against 3D sarcoma spheroids (58% vs 13%; E:T 2:1; 48 h; N = 3, n = 4; p < 0.01). Our findings establish a clinically relevant and translationally valuable platform for evaluating B7-H3-CAR.CIK therapy in dogs with incurable sarcomas, providing a bridge toward the development of novel CAR-based immunotherapies for human incurable sarcomas.
Supplementary Information
The online version contains supplementary material available at 10.1007/s00262-025-04163-4.
Keywords: Canine, CAR, CIK, B7-H3, Sarcoma
Introduction
Advanced-stage sarcomas remain a major therapeutic challenge, often resistant to conventional treatments and in high need for innovative treatment strategies [1, 2]. CAR-engineered lymphocyte therapy has achieved significant success in hematologic malignancies [3, 4] while its clinical translation to sarcomas and other solid tumors still faces significant hurdles [5], underscoring the need for advanced translational research models. In vivo studies, immune deficient mice often fail to capture key aspects of human cancer biology, such as antigen heterogeneity, immunosuppressive microenvironments, and barriers to T cell infiltration [6]. Spontaneously occurring tumors in pet dogs closely resemble human cancers in their biological and clinical complexity, offering an invaluable tool for evaluating CAR-based therapies and bridging the gap to clinical application [7–10]. B7-H3 (CD276) is a promising CAR target due to its broad tumor expression, minimal presence in healthy tissues [11], and association with aggressiveness and stemness features [12]. Its evolutionary conservation between humans and dogs further reinforces its relevance as an experimental target in veterinary oncology [13]. Anti-B7-H3 CAR-T cells have shown potent preclinical antitumor activity, prompting Phase 1 trials in human pediatric solid tumors (NCT04483778). Studies on canine B7-H3 CAR-T cells in osteosarcomas and soft tissue sarcomas highlight both its therapeutic potential in veterinary medicine and its role as a valuable experimental platform [14–16]. Beyond conventional T lymphocytes, alternative immune effectors like CIK (cytokine-induced killer lymphocytes) and NK (natural killer) cells are gaining attention for their intrinsic antitumor activity and potential advantages in solid tumor therapy [17]. CIK cells are ex vivo-expanded lymphocytes generated from peripheral blood mononuclear cells (PBMCs) through sequential culture with IFN-γ, anti-CD3 antibody, and IL-2. This protocol yields a highly proliferative and cytotoxic T cell population capable of potent HLA-independent antitumor activity. Recent studies, from ours and other groups, have shown their potential as a platform for CAR-based strategies against bone and soft tissue sarcomas [18]. In canine patients, although CAR-T cell therapy is under investigation, the feasibility of generating and testing CIK and CAR.CIK remains unexplored.
Given the promising potential of CAR-based immunotherapies in both human and canine cancers, and the emerging and pivotal role of B7-H3 as a shared target between species, this study pursued two primary objectives. First, we generated canine CIK-like lymphocytes using protocols aligned with human studies, ensuring a comparative assessment of CIK killing activity. Second, we investigated the in vitro functional activity of canine B7-H3-CAR.CIK lymphocytes against different canine sarcoma cell lines.
Materials and methods
Immunohistochemistry (IHC)
B7-H3 protein expression was evaluated by immunohistochemistry in 73 formalin-fixed and paraffin-embedded canine tumors [osteosarcoma (n = 22), soft tissue sarcoma (n = 15), hemangiosarcoma (n = 16), and lymphomas (n = 20)]. Following deparaffinization, antigen retrieval was performed using Cell Conditioning Solution 2 (CC2, 950–123, Ventana Medical Systems, Tucson, AZ) at 98 °C for 32 min. Endogenous peroxidase activity was blocked by incubating the sections with DISCOVERY Inhibitor (760–4840, Ventana Medical Systems, Tucson, AZ) at room temperature for 8 min. Subsequently, the primary antibody was manually applied to each slide and incubated at 37 °C for 36 min. This was followed by incubation with the secondary antibody for 16 min.
Sections were incubated with antihuman B7-H3 antibody (Clone RBT-B7-H3, Cat. BSB 2813, BioSB, Santa Barbara, CA) at a dilution of 1:150, and the reaction was performed by Ventana Discovery ULTRA platform (Ventana Medical Systems, Tucson, AZ). Cross-reactivity of the primary antibody was assessed on normal canine tissues as well as on non-neoplastic tissue adjacent to the tumor. As negative controls, we included skin and bone tissue samples from two healthy dogs (one male and one female). B7-H3 expression was scored using a semiquantitative scale (weak, moderate and strong, Suppl. Table 1, 2).
Generation of primary canine sarcoma cell and 3D spheroid models
Primary cell lines were established from surgical specimens of five osteosarcoma (Penny, Wall, Desmon, Nina, and Pedro) and two soft tissue sarcomas (Ivan and Minny); two commercially canine osteosarcoma cell lines, D22 and D17, were included in this study. Tumor cells were isolated from surgical specimens through mechanical dissociation and enzymatic digestion. The enzymatic step was performed using Collagenase Type IV-A (Sigma-Aldrich) at 37 °C for 30 to 60 min. A fibroblast cell line (sample 00#99), was successfully established and characterized by immunocytochemistry for vimentin and α-smooth muscle actin (α-SMA), confirming its mesenchymal identity.
Three-dimensional (3D) sarcoma spheroids were successfully generated (Penny, Wall, and Nina). Detailed characteristics of all cell lines are provided in Suppl. Table 3. 3D sarcoma spheroids were generated as a single spheroid per well using ultra-low attachment (ULA) 96-well round-bottom plates (Corning) with no additional coating. A sarcoma cell suspension of 500–5000 cells/100 mL was plated into wells and then centrifuged at 1.000 g for 10 min [19] Sarcoma spheroids were assembled in 1–4 days, depending on the target histotypes and were maintained under normoxic conditions (37 °C, 21% O2, 5% CO2) in a cell culture incubator.
Western blot
Western blot was performed on protein lysates of Penny, Wall, Ivan, and Minny cell lines. Protein extraction was performed using a lysis buffer composed of 1% Triton X-100, 10% glycerol, 50 mM Tris, 150 mM sodium chloride, 2 mM EDTA (pH 8.0), and 2 mM magnesium chloride, supplemented with a Protease Inhibitor Cocktail (P8340, Sigma-Aldrich). A total of 20 µg of protein from each sample was resolved on a 10% SDS-PAGE gel and subsequently transferred onto a 0.2-µm pore-size nitrocellulose membrane (Thermo Fisher Scientific Inc.). Membranes were blocked with 10% bovine serum albumin (BSA) in Tris-buffered saline (TBS) for 1 h at room temperature and then incubated overnight at 4 °C with the anti-B7-H3 primary (Clone Dkj9M2L mAb, #14,058, Cell Signaling, Danvers, MA) diluted 1:1000 [13] α-tubulin which was used as a loading control. Following primary antibody incubation, membranes were incubated with a horseradish peroxidase (HRP)-conjugated antirabbit secondary antibody (diluted 1:15,000 in TBS-Tween). After six washes in TBS-Tween, signal detection was carried out using Clarity Western ECL Substrate (Bio-Rad Laboratories, Hercules, CA). Protein bands were visualized using CL-XPosure Film (Thermo Fisher Scientific, Waltham, MA), and the resulting images were acquired with an Epson flatbed scanner. MDCK (Madin-Darby Canine Kidney), a normal canine fibroblast culture established in our laboratory (sample ID 00#99), and OVCAR3 cells (human ovarian carcinoma cell line) were used, respectively, as negative and positive controls.
Ex vivo generation and expansion of canine CIK
Canine CIK cells were generated according to the protocol described by Rotolo et al. [20] (STAR Protoc. 2021 Oct 22; 2(4):100,909), which outlines the procedure for T cell activation in dogs. Canine CIK were generated from peripheral blood mononuclear cells (PBMCs) isolated from canine sarcoma-affected canine patients and healthy donors and grown in IMDM 25 mM HEPES (Gibco) medium containing 1,000 U/mL human IFNγ (Miltenyi Biotec) at day 0. The adaptation to the original protocol by pre-stimulating PBMCs with human IFN-γ for 24 h prior to activation was based on an established human CIK generation protocol [18, 19]
After 24 h (day 1), cells were activated using canine anti-CD3 (Biorad) and anti-CD28 mAbs (ThermoFisher) loaded on Dynabeads Tosylactivated (ThermoFisher), with addition of human IL-2 IS (300 U/mL; Miltenyi Biotec). CIK lymphocytes were expanded for three weeks. IL-2 was replenished at every medium change or addition throughout the expansion period. Canine healthy donors were clinically healthy dogs undergoing elective orthopaedic procedures at the veterinary hospital.
Phenotypic validation of CIK cell identity was performed by flow cytometry throughout the expansion period. The resulting population consistently exhibited the phenotype of CIK cells, characterized by CD3 and CD5 (the functional CD3 analog in the canine system), and NKp46 (the canine functional analog of human CD56).
Development of B7-H3-specific CAR-engineered CIK
Retroviral particles encoding chimeric antigen receptors specific for human B7-H3 (B7-H3-CAR) and CD28 costimulatory endodomains were generously provided by Prof. G. Dotti (University of North Carolina) [11]. Canine B7-H3-CAR.CIK lymphocytes were generated by transducing canine PBMCs on day 2 of culture with 1 mL of retroviral supernatant in retronectin-coated plates during overnight incubation. B7-H3-CAR-transduced CIK cells underwent three-week expansion, with IL-2 supplementation (300 U/mL) every 2–3 days, according to established protocols [19]
Multiparameter flow cytometry
Conjugated monoclonal antibodies targeting canine CD3 (clone CA17.2A12, mouse antidog monoclonal Ab, Bio-Rad), CD4 (clone YKIX302.9, rat antidog monoclonal Ab, Bio-Rad), CD8 (clone YCATE55.9, rat antidog monoclonal Ab, Bio-Rad), and CD5 (clone, YKIX322.3, rat antidog monoclonal Ab, Bio-Rad) were used for the characterization of canine B7-H3-CAR.CIK. Additionally, an antihuman NKp46 mAb (Clone 9E2, mouse antihuman monoclonal Ab, BD Pharmingen) was employed. To detect CAR expression, a monoclonal antibody specific for the IgG1/CH2CH3 spacer region (Jackson ImmunoResearch) was used. For sarcoma cell staining, a conjugated antihuman B7-H3 mAb (clone: MIH42, BioLegend) was selected [13]. Labeled cells were acquired using a BD Celesta flow cytometer (BD Pharmingen) and analyzed with FlowJo software.
In vitro assessment of CAR.CIK cytotoxicity against sarcoma
Tumor-killing capacity of canine B7-H3-CAR.CIK lymphocytes and unmodified NTD.CIK lymphocytes was evaluated in vitro against both canine sarcoma cell lines and 3D spheroids using flow cytometry (Celesta, BD Pharmingen); in three cases (Minny, Nina, and Ivan), CAR.CIK and sarcoma cell cultures were derived from samples collected from the same dog, providing an autologous system. Target cells were labeled with the vital dye PKH26 (Sigma-Aldrich) following the manufacturer's protocols. CIK cells were co-cultured for 48 h or 72 h with target cells at various effector-to-target (E:T) ratios (10:1, 5:1, 2.5:1, 1:1, 1:2, and 1:4) in cytotoxicity assays. In selected experiments, we also tested the cytotoxic activity at very low E ratios (1:2, 1:4, and 1:8). The cytotoxic effect was quantified by measuring the residual tumor cells basing on physical parameters, DAPI permeability (DAPI-), and PKH26 staining (PKH26 +). To evaluate spontaneous cell death, target cells were also cultured separately from CIK cells as a control. The percentage of tumor-specific lysis for each E:T ratio was calculated using the following formula: [(experimental mortality-spontaneous mortality)/(100-spontaneous mortality)] × 100.
Spheroid infiltration analysis by confocal microscopy
Sarcoma spheroids were seeded one per well in ultra-low attachment (ULA) 96-well round-bottom plates as detailed in Leuci et al. 2020 and stained with red PKH26 dye [19]. Canine B7-H3-CAR.CIK and unmodified NTD.CIK lymphocytes were stained with green PKH67 dye and plated at an effector to target at E:T ratio 2:1 in culture medium containing 300 U/mL IL-2. After a 12-h co-culture, CIK cells were removed, and immunofluorescence acquisition was performed on the remaining spheroids. Canine sarcoma spheroids were fixed in 4% paraformaldehyde for 1 h and observed using a Leica SP8 AOBS confocal microscope. Image acquisition of the sarcoma spheroids was carried out under consistent settings for laser power, gain, offset, and magnification (20x).
Statistical analysis
All experiments were conducted at least in duplicate to ensure reproducibility. Data analysis was performed using GraphPad Prism 8.0 software (GraphPad Software). Descriptive statistics are presented as mean values ± standard error (SE). For statistical comparisons between two groups, two-tailed Student's t-tests were employed. For comparisons involving three or more groups, two-way analysis of variance (ANOVA) was used, followed by Bonferroni test. Statistical significance was set at P < 0.05.
Results
B7-H3 is highly expressed by multiple canine sarcoma subtypes
B7-H3 was found to be highly expressed in 74% of the analyzed canine sarcoma samples (39/53): 68% (15/22) of osteosarcomas, 93% (14/15) of soft tissue sarcomas, and 63% (10/16) of hemangiosarcoma. Representative images of B7-H3 expression in canine tumors are shown in Fig. 1a–d. Canine soft tissue sarcomas show strong B7-H3 expression, whereas surrounding normal tissues and normal canine tissues demonstrate no immunoreactivity (Fig. 1e–g; Suppl. Table 1, 2).
Fig. 1.
B7-H3 expression in canine sarcoma samples and cell lines. Representative expression of B7-H3 in canine osteosarcomas a, soft tissue sarcomas b, and hemangiosarcoma c samples. d Negative versus low score positivity of B7-H3 in lymphoma. No reactivity in e muscle tissue margin, f skin tissue margin, g adipose tissue margin of canine soft tissues sarcomas. All images were captured at 10X magnification. h Workflow for generating primary sarcoma cell lines from surgical specimens with concurrent analysis of tumor-infiltrating lymphocytes. i Flow cytometric assessment of B7-H3 surface expression across sarcoma cell lines, with MDCK cells serving as a B7-H3-negative control. Isotype controls depicted in blue. l Validation of B7-H3 protein expression by Western blot analysis in selected sarcoma cell lines (Penny, Wall, Ivan, Minny) with OVCAR3 and both MDCK and normal canine fibroblast 00#99as positive and negative controls, respectively, normalized to tubulin
Primary cell lines were established from surgical specimens obtained from nine canine tumor patients: seven with osteosarcoma (Penny, Wall, Desmon, Nina, and Pedro, D17 and D22) and two with soft tissue sarcoma (Ivan and Minny) (Fig. 1h; Suppl.Table 3). All the analyzed canine sarcoma cell lines, regardless of histological subtype, were positive for B7-H3 expression, although with heterogeneous levels of surface expression. Representative histograms illustrating the variable intensity of B7-H3 expression across different sarcoma cell lines are shown in Fig. 1i (N = 9). In selected cases, B7-H3 expression was further confirmed by Western blot analysis. We compared B7-H3 expression in canine and human tumors (OVCAR3), confirming the cross reactivity of the human anti-B7-H3 antibody by Western blot. Both MDCK cell lines and normal canine fibroblast (00#99) were negative for B7-H3 expression (Fig. 1l.
Generation and functional characterization of canine B7-H3-CAR.CIK lymphocytes
Canine CIK cells were successfully generated from PBMCs collected from five dogs, three diagnosed with sarcoma (osteosarcoma n = 1, fibrosarcoma n = 1, and soft tissue sarcoma n = 1) and two healthy donors. CIK cells were then genetically engineered with a second-generation human anti-B7-H3-CAR, incorporating the CD28 costimulatory domain (Fig. 2a). Two weeks post-transduction and ex vivo expansion, the mean CAR expression was 39% ± 2% on the engineered CIK cells (Table 1, Fig. 2b). The ex vivo expansion (fivefold, range 2–13) and immunophenotype (CD3: 92% ± 3%; CD8: 87% ± 7%; CD4: 49% ± 5%; CD5: 77% ± 0.1%; NKp46: 83% ± 5%; N = 5) of B7-H3-CAR.CIK lymphocytes were not impaired by the CAR engineering procedures, as they were comparable to those of unmodified NTD.CIK lymphocytes.
Fig. 2.
Engineering and characterization of canine B7-H3-CAR.CIK lymphocytes. a Generation of B7-H3-CAR.CIK lymphocytes from canine PBMCs. b Representative flow cytometry plots showing CAR expression on canine CIK
Table 1.
Comprehensive immunophenotype characterization comparing canine B7-H3-CAR.CIK and unmodified CIK lymphocytes (NTD.CIK)
| CD3 (%) | CAR (%) | CD4 (%) | CD8 (%) | CD5 (%) | NKp46 (%) | |
|---|---|---|---|---|---|---|
| CAR.CIK | 92.4 ± 3.4 | 38.6 ± 1.9 | 48.6 ± 4.7 | 87.2 ± 7.1 | 77 ± 0.1 | 82.8 ± 4.9 |
| NTD.CIK | 81.8 ± 7.2 | / | 42 ± 5.9 | 73.8 ± 10.7 | 70 ± 7.0 | 81 ± 5.9 |
B7-H3-CAR.CIK lymphocytes effectively kill canine sarcomas
We developed a pet dog-derived experimental platform using CIK lymphocytes, sarcoma cell cultures, and 3D spheroids (Fig. 3a). Canine B7-H3-CAR.CIK effectively eliminated canine sarcomas in vitro, demonstrating significantly higher cytotoxicity compared to NTD.CIK (45% ± 6% vs. 8% ± 3%; E:T 1:1, 48 h; N = 7, n = 8; p < 0.0001; Fig. 3b; Suppl. Figure 1a). The antitumor activity of canine B7-H3-CAR.CIK lymphocytes remained evident at very low E:T ratios (31% ± 6% vs. 2% ± 1%; E:T 1:4, 72 h; N = 7, n = 8; p < 0.0001; Fig. 3c). Of note, in 3/7 assays, we could replicate an autologous setting with canine B7-H3-CAR.CIK and sarcoma cell cultures generated from the same dog without differences in killing efficacy compared to non-autologous setting (Fig. 3d, Suppl. Figure 1b). At the E:T ratios that produced the most potent antitumor effects, canine B7-H3-CAR.CIK lymphocytes selectively spared canine normal cells (MDCK cell line), which do not express B7-H3 (n = 3; Fig. 3e; Suppl. Figure 1c). To further explore canine B7-H3-CAR.CIK killing kinetics, we developed 3D sarcoma spheroids (n = 3) that more closely mimic the solid tumor architecture. Canine B7-H3-CAR.CIK lymphocytes demonstrated active tumor infiltration (monitored after 12 h of co-culture) and killing activity against 3D canine sarcoma spheroids, significantly outperforming unmodified NTD.CIK (58% ± 4% vs. 13% ± 6%; E:T 2:1, 48 h; N = 3, n = 4; p < 0.01; Fig. 4a–d) CAR.CIK.
Fig. 3.
B7-H3-CAR.CIK lymphocytes cytotoxicity against canine sarcoma cell lines. a Overview of the comprehensive preclinical testing platform. Superior cytotoxic activity of B7-H3-CAR.CIK compared to unmodified CIK lymphocytes (NTD.CIK) at standard b or low c E:T ratios. d Effective activity by B7-H3 CAR.CIK against selected autologous sarcoma cells. e Minimal activity by B7-H3 CAR.CIK against B7-H3-negative MDCK cells (n = 3). Statistical analysis performed using two-way ANOVA with Bonferroni correction; significance levels: *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001
Fig. 4.
Canine B7-H3-CAR.CIK lymphocytes activity against three-dimensional canine sarcoma models. a Comparative bright-field microscopy of canine sarcoma spheroids treated with NTD.CIK, B7-H3-CAR.CIK or untreated. b Flow cytometry analysis quantifying B7-H3-CAR.CIK lymphocytes-mediated cytotoxicity against 3D sarcoma spheroids at E:T ratio 2:1. c Enhanced spheroid disruption by B7-H3-CAR.CIK lymphocytes compared to unmodified CIK cells. d Confocal microscopy maximum intensity projections showing differential infiltration patterns of NTD.CIK cells versus B7-H3-CAR.CIK lymphocytes (PKH67-stained, green) in sarcoma spheroids (PKH26-stained, red) at E:T 2:1
Discussion
In this study, we developed a preclinical model of CAR-based cellular immunotherapy using canine lymphocytes, which are comparable to human CIK cells in their generation and functional characteristics. We confirmed that B7-H3 is a tumor-associated target broadly expressed across multiple sarcoma histotypes, highlighting its significant clinical relevance for both veterinary applications and translational research toward human clinical studies.
Our data on B7-H3, both in terms of its wide tumor expression and limited presence on normal cells, confirm the strong homology between species and underscore the translational relevance of conducting cellular immunotherapy research in the veterinary field.
Our study represents the first demonstration of generating canine CIK and CAR.CIK exploiting an ex vivo expansion protocol adapted from its human counterpart. The resulting cellular product is comparable in quantitative, functional, and phenotypic terms to what has been reported for human CIK generation [18, 19]. It should also be noted that the NKp46 antibody employed in this study is of human origin, and although reported to cross-react with canine cells, its binding specificity across species cannot be conclusively assumed. Further dedicated validation will therefore be required to precisely define NK-like phenotypic features in canine CIK cells. Phenotypic characterization of CD4 and CD8 expression was performed using single-color staining to minimize inaccuracies related to the limited specificity of available canine reagents. Although this approach may introduce minor imprecision, the averaged values still indicate the presence of a small CD4⁺CD8⁺ double-positive subset, whose biological relevance in peripheral canine lymphocytes remains to be clarified. For CAR-transduced lymphocytes, positivity was defined relative to a negative control sample. We acknowledge that this threshold is inherently arbitrary and may underestimate cells with low-level receptor expression, it was conservatively chosen to report CAR expression within heterogeneously transduced populations. The robust ex vivo expansion observed is a critical requirement for clinical applications, particularly when envisioning protocols starting from limited volumes of peripheral blood, targeting an ideal therapeutic dose ranging from 106 to 107 lymphocytes/kg.
The in vitro killing activity of canine anti-B7-H3 CAR.CIK, reproducible across multiple canine sarcoma histologies, aligns closely with previous findings in the human setting. It has to be considered that CIK possesses limited intrinsic HLA-independent antitumor potential, which, when enhanced by CAR specificity, enables efficacy even at low E:T ratios, representative of clinical conditions where unmodified CIK cells would be ineffective.
Furthermore, observations from 3D tumor spheroid models support the properties and potential of this approach, demonstrating the ability to penetrate and effectively target inner layers of tumor structures. Additionally, a CAR engineering platform based on CIK cells offers the possibility of addressing tumors with heterogeneous CAR target expression, extending the versatility and applicability of this therapeutic strategy.
Limitations of this study include the restricted representation of canine sarcomas, limited to osteosarcoma, fibrosarcoma, and soft tissue sarcoma. This narrow focus does not encompass the full histological diversity observed in human sarcomas. Accordingly, it is important to interpret the findings considering interspecies differences in subtype prevalence, epidemiology, and molecular features, all of which may influence the translational relevance of the results. In this context, the canine model should be viewed as complementary to murine systems, offering unique advantages, particularly for the study of spontaneously arising, immunocompetent tumors. Future studies should expand the spectrum of included sarcoma subtypes and deepen comparative molecular analyses to better assess cross-species relevance and strengthen translational insights.
Further limitation lies in the use of a human-derived anti‑B7‑H3 CAR construct, which, while functionally cross‑reactive in canine cells, may not fully recapitulate the optimal design for veterinary applications. Efforts are currently underway to develop a fully canine-specific anti-B7-H3 CAR for future clinical veterinary studies.
While promising, the current in vitro results may not fully predict in vivo efficacy due to the complexity of the tumor microenvironment and potential immune suppression mechanisms. The long-term persistence and functionality of these cells in vivo require further investigation through longitudinal studies. Additionally, although B7-H3 expression was not detected in selected normal tissues, potential off-tumor toxicity must be carefully evaluated in future clinical applications.
Overall, our findings provide a robust rationale for supporting the promotion and development of clinical investigational approaches using anti-B7-H3-CAR.CIK cellular immunotherapy in the veterinary field, particularly against inoperable sarcomas. This strategy holds dual significance: offering immediate therapeutic benefits for pet dogs while also serving as a highly informative model for designing similar studies in human patients with currently incurable sarcomas, overcoming the limitations of in vivo experimental models using immunocompromised animals.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
The authors thank the Fondazione Piemontese per la Ricerca sul Cancro (FPRC), Institute of Cancer research (IRCC) of Candiolo and Associazione Italiana Ricerca sul Cancro (AIRC) for funding and research supports.
Author contributions
R.D.M., C.D.: conceptualization, formal analysis, supervision, investigation, visualization, writing-original draft, writing-review and editing. S.C., A.M., F.G., A.P., L.V., L.C.C., P.A., A.S., G.G, E.B: investigation, writing-review and editing. G.D., E.L., E.V., V.L.: conceptualization, writing-review and editing. L.A., D.S.: conceptualization, resources, supervision, funding acquisition, writing-original draft, writing-review and editing.
Funding
This work was funded by Fondazione Piemontese per la Ricerca sul Cancro (FPRC) (5X1000) (L.Aresu), Institute of Cancer Research (IRCC) of Candiolo (Ricerca Corrente 2024-2025) (D.Sangiolo) and Associazione Italiana Ricerca sul Cancro (AIRC) IG-2017 N. 20259 (D.Sangiolo) and N.23104 (G.Grignani)
Data availability
No datasets were generated or analysed during the current study.
Declarations
Conflict of interest
The authors declare no conflict of interest.
Footnotes
Raffaella De Maria and Chiara Donini these are the co-first authors .
Dario Sangiolo and Luca Aresu these are the co-last authors .
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Change history
12/23/2025
A Correction to this paper has been published: 10.1007/s00262-025-04220-y
References
- 1.Pollack SM, Ingham M, Spraker MB, Schwartz GK (2018) Emerging targeted and immune-based therapies in sarcoma. J Clin Oncol 36(2):125–135. 10.1200/jco.2017.75.1610 [DOI] [PubMed] [Google Scholar]
- 2.Pasquali S, Pizzamiglio S, Touati N, Litiere S, Marreaud S, Kasper B et al (2019) The impact of chemotherapy on survival of patients with extremity and trunk wall soft tissue sarcoma: revisiting the results of the EORTC-STBSG 62931 randomised trial. Eur J Cancer 109:51–60. 10.1016/j.ejca.2018.12.009 [DOI] [PubMed] [Google Scholar]
- 3.Grupp SA, Kalos M, Barrett D, Aplenc R, Porter DL, Rheingold SR et al (2013) Chimeric antigen receptor-modified T cells for acute lymphoid leukemia. N Engl J Med 368(16):1509–1518. 10.1056/NEJMoa1215134 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Geldres C, Savoldo B, Dotti G (2016) Chimeric antigen receptors for cancer immunotherapy. Methods Mol Biol 1393:75–86. 10.1007/978-1-4939-3338-9_7 [DOI] [PubMed] [Google Scholar]
- 5.Pollack SM, Loggers ET, Rodler ET, Yee C, Jones RL (2011) Immune-based therapies for sarcoma. Sarcoma 2011:438940. 10.1155/2011/438940 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Buqué A, Galluzzi L (2018) Modeling tumor immunology and immunotherapy in mice. Trends Cancer 4(9):599–601. 10.1016/j.trecan.2018.07.003 [DOI] [PubMed] [Google Scholar]
- 7.Sakai O, Igase M, Mizuno T (2020) Optimization of canine CD20 chimeric antigen receptor T cell manufacturing and in vitro cytotoxic activity against B-cell lymphoma. Vet Comp Oncol 18(4):739–752. 10.1111/vco.12602 [DOI] [PubMed] [Google Scholar]
- 8.Paoloni M, Khanna C (2008) Translation of new cancer treatments from pet dogs to humans. Nat Rev Cancer 8(2):147–156. 10.1038/nrc2273 [DOI] [PubMed] [Google Scholar]
- 9.Mata M, Vera JF, Gerken C, Rooney CM, Miller T, Pfent C et al (2014) Toward immunotherapy with redirected T cells in a large animal model: ex vivo activation, expansion, and genetic modification of canine T cells. J Immunother 37(8):407–415. 10.1097/cji.0000000000000052 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Yoshimoto S, Kudo A, Rotolo A, Foos K, Olenick L, Takagi S et al (2024) Validation of a PD-1/CD28 chimeric switch receptor to augment CAR-T function in dogs with spontaneous B cell lymphoma. iScience 27(9):110863. 10.1016/j.isci.2024.110863 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Du H, Hirabayashi K, Ahn S, Kren NP, Montgomery SA, Wang X et al (2019) Antitumor responses in the absence of toxicity in solid tumors by targeting B7–H3 via chimeric antigen receptor T cells. Cancer Cell 35(2):221–37.e8. 10.1016/j.ccell.2019.01.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Zhao B, Li H, Xia Y, Wang Y, Shi Y, Xing H et al (2022) Immune checkpoint of B7–H3 in cancer: from immunology to clinical immunotherapy. J Hematol Oncol 15(1):153. 10.1186/s13045-022-01364-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Zhang S, Black RG, Kohli K, Hayes BJ, Miller C, Koehne A et al (2022) B7–H3 specific CAR T cells for the naturally occurring, spontaneous canine sarcoma model. Mol Cancer Ther 21(6):999–1009. 10.1158/1535-7163.mct-21-0726 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Cao JW, Lake J, Impastato R, Chow L, Perez L, Chubb L et al (2024) Targeting osteosarcoma with canine B7–H3 CAR T cells and impact of CXCR2 co-expression on functional activity. Cancer Immunol Immunother 73(5):77. 10.1007/s00262-024-03642-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Lynch MM, Al-Marayaty R, Obeidin F, Alexiev BA, Chen EY, Viveiros P et al (2024) B7–H3 is widely expressed in soft tissue sarcomas. BMC Cancer 24(1):1336. 10.1186/s12885-024-13061-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Wang L, Zhang Q, Chen W, Shan B, Ding Y, Zhang G et al (2013) B7–H3 is overexpressed in patients suffering osteosarcoma and associated with tumor aggressiveness and metastasis. PLoS ONE 8(8):e70689. 10.1371/journal.pone.0070689 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Sangiolo D (2011) Cytokine induced killer cells as promising immunotherapy for solid tumors. J Cancer 2:363–368. 10.7150/jca.2.363 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sangiolo D, Mesiano G, Gammaitoni L, Aglietta M, Grignani G (2014) Activity of cytokine-induced killer cells against bone and soft tissue sarcoma. Oncoimmunology 3:e28269. 10.4161/onci.28269 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Leuci V, Donini C, Grignani G, Rotolo R, Mesiano G, Fiorino E et al (2020) CSPG4-specific CAR.CIK lymphocytes as a novel therapy for the treatment of multiple soft-tissue sarcoma histotypes. Clin Cancer Res 26(23):6321–6334. 10.1158/1078-0432.ccr-20-0357 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Rotolo A, Atherton MJ, Kasper BT, Haran KP, Mason NJ (2021) Genetic re-direction of canine primary T cells for clinical trial use in pet dogs with spontaneous cancer. STAR Protoc 2(4):100905. 10.1016/j.xpro.2021.100905 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
No datasets were generated or analysed during the current study.