FAP-CD40 and PD1-IL2v combination therapy reprograms immunologically cold tumors through de novo intratumoral T cell-dendritic cell clusters.

FAP-CD40 and PD1-IL2v combination therapy reprograms immunologically cold tumors through de novo intratumoral T cell-dendritic cell clusters - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Search Log in Primary site navigation Search Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice J Immunother Cancer . 2026 May 28;14(5):e014620. doi: 10.1136/jitc-2025-014620 FAP-CD40 and PD1-IL2v combination therapy reprograms immunologically cold tumors through de novo intratumoral T cell-dendritic cell clusters Thuy Trinh Nguyen 1 , Harold Gómez 2 , Mechthild Lutge 1 , Emilio Yángüez 1 , Tamara Hüsser 1 , Sina Nassiri 2 , Christine Trumpfheller 1 , Sara Colombetti 2 , Laura Codarri Deak 1 , Pablo Umaña 1 , Sònia Tugues 3, 4 , Ines Grazina de Matos 1 , Leo Kunz 2, ✉ Author information Article notes Copyright and License information 1 Roche Pharma Research and Early Development, Roche Innovation Center Zurich, Schlieren, Switzerland 2 Roche Pharma Research and Early Development, Roche Innovation Center Basel, Basel, Switzerland 3 Institute of Experimental Immunology, Universität Zürich, Zürich, Switzerland 4 Department of Immunology, Heidelberg University Medical Faculty Mannheim, Mannheim, Germany Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise. Supplement: Additional supplemental material is published online only. To view, please visit the journal online ( https://doi.org/10.1136/jitc-2025-014620 ). FAP-CD40 and PD1-IL2v are proprietary molecules developed by F Hoffmann-La Roche. CT, SC, LCD and PU are inventors on patent applications covering these molecules. TTN, HG, ML, EY, TH, SN, CT, SC, LCD, PU, IGM and LK were employees of Roche at the time of the study. HG, EY, TH, SN, CT, SC, LCD, PU, IGM and LK declare ownership of Roche stock. No disclosures were reported by the other authors. ✉ Dr Leo Kunz; leo.kunz@roche.com Received 2025 Dec 15; Accepted 2026 May 14; Collection date 2026. Copyright © Author(s) (or their employer(s)) 2026. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ Group. This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See https://creativecommons.org/licenses/by-nc/4.0/ . PMC Copyright notice PMCID: PMC13223653  PMID: 42208978 Abstract Background Pancreatic ductal adenocarcinoma (PDAC) remains a major challenge for immunotherapy due to its immunologically cold tumor nature, characterized by poor T cell infiltration and a highly suppressive tumor microenvironment. Here, we propose a novel strategy, combining fibroblast activation protein (FAP)-CD40 to activate dendritic cells (DCs) in the tumor microenvironment and programmed cell death protein-1 (PD1)-interleukin 2v (IL2v) to promote the expansion and differentiation of tumor-infiltrating T cells. We hypothesize that this combination will synergistically enhance both T cell priming and expansion directly within pancreatic 4662 KPC tumors, which recapitulate the immunologically cold features of human PDAC. Methods Immune cell distribution and abundance following FAP-CD40/PD1-IL2v monotherapy or combination therapy were analyzed using multiplexed confocal imaging (3D immune phenotyping). FTY720 studies assessed the contribution of lymph node priming in treatment efficacy, while CD4+/CD8+ T cell depletion experiments identified the roles of these subsets in combination therapy. T cell functionality was further assessed through ex vivo restimulation assays and single-cell RNA sequencing. Results Combination therapy induced dense intratumoral clusters of CD4 + and CD8 + T cells, colocalized with type 1 conventional DCs, termed as T cell-DC clusters (TDCs). These TDCs were strongly associated with tumor regression, which required both CD4 + and CD8 + T cells. Furthermore, T cells from combination-treated tumors showed enhanced functionality, with increased tumor necrosis factor-alpha and interferon-gamma production compared with monotherapy groups. Single-cell RNA sequencing revealed polarization of CD4 + T cells toward a T helper cell 1 phenotype in combination-treated tumors. Conclusion The combination of FAP-CD40 and PD1-IL2v offers a promising strategy for treating poorly infiltrated, cold tumors. By driving T cell infiltration, promoting de novo TDC formation and orchestrating local antitumor immunity, this strategy provides a foundation for future therapies targeting immunotherapy-resistant tumors. Keywords: Immunotherapy, Solid tumor, Tumor microenvironment - TME, Tumor infiltrating lymphocyte - TIL, Combination therapy WHAT IS ALREADY KNOWN ON THIS TOPIC Immunotherapies, such as checkpoint inhibitors, have transformed cancer treatment, however, their efficacy is largely restricted to tumors with an inflamed tumor microenvironment (TME). In contrast, immunologically cold tumors—such as pancreatic ductal adenocarcinoma (PDAC)—exhibit low levels of T cell infiltration, with T cells often excluded from the tumor parenchyma. This lack of T cell infiltration is a key factor underlying the limited success of current immunotherapies in treating cold tumors. WHAT THIS STUDY ADDS The combination of fibroblast activation protein (FAP)-CD40 and programmed cell death protein-1 (PD1)-interleukin 2v (IL2v) induces significant synergistic tumor regression in the immunologically cold PDAC model. This therapy promotes the de novo formation of T cell-dendritic cell (DC) clusters (TDCs), which are strongly associated with treatment response. TDCs serve as local immune hubs, comprising type 1 conventional DC-CD4 + /CD8 + T cell aggregates, and function independently of continuous T cell priming in the lymph nodes. The therapeutic effect depends on the simultaneous presence of both CD4 + and CD8 + T cell subsets at the onset of treatment. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY This study presents a therapeutic strategy to transform cold tumors by promoting the formation of organized immune structures (TDCs) within the TME. The FAP-CD40 and PD1-IL2v combination provides a promising approach for addressing immunotherapy-resistant solid tumors. These findings highlight the need for strategies that enhance spatial organization and localized activation of immune cells within the TME, offering potential advancements in cancer immunotherapy. Background Immunotherapies have revolutionized cancer treatment by attempting to harness the patients’ immune system in the fight against cancer. Cancer immunotherapy encompasses various strategies including adoptive cell therapies, cancer vaccines, targeted cytokine treatments and immune checkpoint inhibitors. 1 , 4 Among these strategies, immune checkpoint inhibition has shown clinical benefits and improved survival rates in patients whose tumors have previously resisted conventional chemotherapy regimens. 5 Checkpoint blocking antibodies disrupt inhibitory immune signaling pathways, such as the programmed cell death protein-1 (PD-1)/programmed death-ligand 1 or cytotoxic T-lymphocyte-associated protein 4 signaling axes, releasing tumor-reactive T cells from immunosuppression. 6 However, these agents demonstrate efficacy predominantly in cancers characterized by an inflamed tumor microenvironment (TME), such as melanoma and non-small cell lung cancer. 7 In contrast, immunologically cold tumors such as pancreatic ductal adenocarcinoma (PDAC), show poor T cell infiltration and limited responsiveness to checkpoint blockade therapy. 8 Transforming the cold TME of pancreatic and other tumors into an inflamed state remains a critical step in immunotherapeutic interventions. 9 In cold tumors, the few T cells present predominantly reside at the tumor periphery or within stromal compartments rather than infiltrating the tumor parenchyma. 10 11 This state of immune cell exclusion is often exacerbated by a deficiency in conventional type 1 dendritic cells (cDC1s), which play essential roles in cross-presenting tumor antigens and orchestrating further T cell recruitment to the tumor site. 12 , 14 Previous studies in pancreatic KPC mouse tumor models (KRAS(G12D) and Trp53 mutations), which exhibit a fibrotic and immunologically cold phenotype mirroring human PDAC, 15 16 despite lower T cell density in mice compared with humans, 17 have shown that CD40 agonism can rescue cDC1 dysfunction and promote its activation. 12 However, this success was often accompanied by a reduction in intratumoral cDC1 numbers, 12 18 suggesting their migration to tumor-draining lymph nodes (tdLNs) and a consequent reduction in interactions with intratumoral T cells. Combining CD40 agonism with Fms-related tyrosine kinase 3 ligand successfully restored intratumoral cDC1 numbers and enhanced T cell recruitment to the tumor. 14 18 Despite these improvements, neither approach elicited a strong antitumor response, despite evidence of T cell priming in tdLNs. 14 Another strategy involves activating dendritic cells (DCs) specifically within the TME using fibroblast activation protein (FAP)-CD40, which triggers CD40 stimulation only in the presence of the FAP on cancer-associated fibroblasts. FAP-CD40 enables localized DC activation within the TME, thereby avoiding systemic CD40 activation. 19 This approach showed similar outcomes, with FAP-CD40 leading to moderate tumor growth inhibition of KPC tumors but achieved complete regression in the inflamed MC38 colon cancer model. 19 These findings suggest that lymph node (LN) priming alone might be insufficient to drive robust antitumor immunity in the cold pancreatic model, and infiltrating T cells likely require additional signals within the TME to overcome local immunosuppression to achieve tumor regression. 20 A potential strategy to provide T cells in the TME with additional stimuli to induce tumor regression involves PD1-interleukin 2v (IL2v), an immune-targeted cytokine fusion protein that selectively delivers a mutated IL-2, devoid of CD25 binding, to PD-1-expressing T cells. 21 By preventing CD25 binding, PD1-IL2v prevents activation of regulatory T cells and instead drives the expansion and differentiation of antigen-experienced TCF1 + PD-1 + stem-like CD8 + T cells into effector T cells. While PD1-IL2v enhances T cell functionality and has shown significant tumor growth control in B16-F10-OVA tumor models, its effectiveness may be limited in tumors with insufficient T cell priming and a low abundance of antigen-specific T cells. 21 Moreover, we hypothesize that combining FAP-CD40 with PD1-IL2v will synergistically enhance both T cell priming and expansion directly within the tumor, increasing the likelihood of localized interactions between DCs and T cells within the TME. Proper spatial positioning of DCs and T cells within the TME is crucial, 22 as recent advances in spatial immunoprofiling have shown that spatially organized structures within the TME are key determinants of therapeutic success. Specifically, these studies have identified the presence of intratumoral immune hubs, such as antigen-presenting cell niches or “immune triads”, 23 , 25 where DCs and T cells colocalize and interact, which are strongly associated with effective anticancer immunity. While these spatially organized immune hubs have been characterized in immunologically hot tumors, approaches to induce their formation in cold tumors remain largely unexplored. Moreover, these hubs were characterized as intrinsic features of the TME, with their presence predicting responsiveness to immune checkpoint blockade. 26 27 Given the lack of strategies to actively induce these structures, we developed a therapeutic approach to promote the formation of functional immune hubs in the TME of non-inflamed tumors ultimately leading to tumor regression in a difficult-to-treat mouse model of PDAC. Using 3D immune phenotyping (3DIP), a platform that combines multiplexed confocal imaging with spatial analysis, 28 we discovered that the combination of FAP-CD40 with PD1-IL2v (here referred to as “combo treatment”) induced specialized immune aggregates composed of CD4 + and CD8 + T cells colocalizing with cDC1s, which we termed “T cell-DC clusters” (TDCs). The presence of TDCs strongly correlated with tumor regression. Functional studies showed that both CD4 + and CD8 + T cells are required for tumor regression, while TDC formation occurred independently of continuous lymphoid priming, demonstrating a self-sustaining intratumoral program. This represents a novel strategy to convert immunologically cold tumors by driving de novo induction of intratumoral TDCs that orchestrate local antitumor immunity. Methods Mice Female C57BL/6J mice (Charles River) or C57BL/6-Tg(Nr4a1-EGFP/cre)820Khog/J (Jackson), 6–8 weeks of age at experiment initiation, were used for animal experimentation. Animals were housed in specific-pathogen-free conditions with 12-hour light/dark cycles. All animal care and experimental procedures were conducted in accordance with institutional guidelines (GV Solas; FELASA; TierschG). Research licenses were reviewed and approved by the Cantonal Veterinary Authorities in Zürich (licenses ZH181/2020, ZH183/2020, ZH184/2020 and ZH001/2024). Based on these approvals, specific study protocols defining the research question, experimental design and analysis were prepared prior to study initiation in accordance with these licenses and reviewed internally. On arrival, mice underwent a 1-week acclimation phase. Subsequently, a subcutaneous microchip was implanted for individual animal identification, followed by an additional 1-week recovery period before the initiation of experiments. Tumor measurements were performed 1–3 times weekly (depending on tumor size) using digital calipers. Tumor volume was calculated using the formula: 0.5 × tumor length × tumor width 2 , as previously described. 29 30 Animal health status was monitored daily, with humane endpoints applied according to predefined termination criteria, including arched back, abdominal distention, decreased activity, tumor volume >2000 mm 3 , body weight loss >20% (license dependent) or ulceration. Adverse events were reported according to the scoring system outlined in the relevant license. Body weight was assessed weekly, with increased monitoring implemented for animals showing weight loss. Predefined inclusion and exclusion criteria based on health status, humane endpoints, and tumor measurements were applied, and all excluded animals were documented and reported. Cell lines The 4662 PDAC mouse cell line was derived from a KrasLSL-G12D/+; Trp53LSL-R172H/+; Pdx1-Cre (KPC) mouse 15 and obtained from Professor Robert H Vonderheide at the University of Pennsylvania. The KPC cell line was cultured in Dulbecco’s Modified Eagle Medium (Thermo Fisher, #42430025) supplemented with 10% fetal bovine serum (FBS) (Sigma-Aldrich, #F4135). Cultures were incubated at 37°C in a humidified atmosphere containing 5% CO 2 , with cells passaged at 70–80% confluence using standard procedures. Tumor cell injection For tumor cell inoculation, KPC cells were dissociated using TrypLE (Thermo Fisher, #12605-010). The cells were then resuspended at a concentration of 3×10 6 /mL in a mixture composed of 50% RPMI 1640 (Gibco, #42401-018) and 50% Matrigel (Corning, #354230). The cell suspension (100 µl/mouse) was subsequently injected into the left flank of anesthetized C57BL/6J mice. FAP-CD40 and PD1-IL2v treatment in vivo Therapy injection was initiated when tumors reached an average volume of 200–250 mm 3 . Mice were randomized into treatment groups either 1 day before or on the day of the first treatment administration to achieve comparable mean tumor volumes across groups without experimenter blinding and without explicit control of potential confounders. Both molecules were formulated in Histidine buffer (Bichsel, customized formulation for Roche Glycart AG) on the day of administration. FAP-CD40 was administered at a dose of 13.3 mg/kg via intraperitoneal injection (200 µl), while PD1-IL2v was delivered intravenously (200 µl) at a dose of 0.5 mg/kg. The molecules were produced and purified at Roche Innovation Center Zurich and Roche Innovation Center Munich. Tumors were collected for downstream analysis on day 10 and at the time of study termination, with mice anesthetized prior to tumor collection. FTY720 treatment in vivo FTY720 (Sigma-Aldrich, #SML0700-25MG) was initially resuspended at a concentration of 10 mg/mL for endotoxin testing. This stock was then diluted to a final dosing concentration of 5 mg/kg body weight. FTY720 was administered via intraperitoneal injection (200 µl) starting 2 days prior to FAP-CD40 and PD1-IL2v treatment start and continued every 4 days throughout the study until termination. To confirm successful depletion of peripheral T cells, blood samples were collected 1 day before the start of treatment and 5 days prior to study termination and subsequently analyzed by Cytek Aurora. CD4 + and CD8 + T cell depletion in vivo For CD4 + and/or CD8 + T cell depletion, anti-mouse CD4 (clone “GK1.5”, BioLegend, #100470) and anti-mouse CD8α (clone “2.43”, Selleck Chemicals, #A2102-5MG*20) antibodies were diluted in Histidine buffer to achieve a final concentration of 5 mg/kg. Mice received intraperitoneal injections of 100 µg (200 µl/mouse) of either anti-CD4 + and/or anti-CD8 + T cell antibody on days 6, 9 and 12 post-tumor inoculation. Depletion of CD4 + and/or CD8 + T cells was confirmed by Cytek analysis of blood samples collected on day 12 (1 day prior to treatment initiation). Processing of blood samples To evaluate T cell depletion efficacy in peripheral blood following either FTY7220 treatment or administration of anti-CD4 and/or anti-CD8 antibodies for T cell depletion, 50 µl of blood was collected from the tail vein of mice into lithium heparin-coated tubes (Sarstedt, #41.1503.015). From each sample, 35 µl was transferred to a U-bottom 96-well plate and treated with BD Pharm Lyse buffer (BD Biosciences, #555899) to lyse the red blood cells. The plate was incubated for 10 min at 4°C, followed by centrifugation and a second lysis step. After, cells were stained with Near-IR Live/Dead viability dye (1:1000, Thermo Fisher, # L34976 ) and CD16/CD32 Fc receptor blocking antibody (1:100, BD Biosciences, #553142) in phosphate-buffered saline (PBS) for 20 min at 4°C. After two washing steps, cell surface staining was performed in PBS for 30 min at 4°C. Following surface staining, cells were washed twice and fixed with BD Cytofix buffer (BD Biosciences, #554655) for 30 min at room temperature. After two additional washing steps, cells were resuspended in FACS buffer (prepared in-house, consisting of deionized H 2 O, PBS, FBS, EDTA at pH 8 and sodium azide) for flow cytometric analysis. Data were acquired on a Cytek Aurora and analyzed using FlowJo software (V.10.10.0) The following antibodies were used for cell surface staining after FTY720 treatment: TCF1 PE (1:200, clone “S33-966”, BD Biosciences, #564217), TCRβ BV570 (1:100, clone “H57-597”, BioLegend, #109231), CD4 BV750 (1:800, clone “RM4.5”, BD Biosciences, #747344), PD-1 BV785 (1:100, clone “29F.1A12”, BioLegend, #135225), CD8 BUV496 (1:400, clone “53-6.7”, BD Biosciences, #750024), CD19 BUV563 (1:800, clone “1D3”, BD Biosciences, 749028) and CD45 BUV805 (1:800, clone “30-F11”, BD Biosciences, 748370). The following antibodies were used for cell surface staining after CD4 + and/or CD8 + T cell depletion: CD8 FITC (1:200, clone “CT-CD8α”, Thermo Fisher, #MA5-17598), NK1.1 PE Cy7 (1:800, clone “PK136”, BioLegend, #108714), CD11c PE Fire810 (1:1600, clone “QA18A72”, BioLegend, #161106), XCR1 AF647 (1:300, clone “ZET”, BioLegend, #148214), CD4 BV421 (1:400, clone “RM4-4”, BD Biosciences, #740008), MHCII ef450 (1:800, clone “M5/114.15.2”, Thermo Fisher, #48-5321-82), TCRβ BV510 (1:100, clone “H57-597”, BD Biosciences, #563221), CD11b BV650 (1:800, clone “M1/70”, BioLegend, #101259), CD19 BUV563 (1:800, clone “1D3”, BioLegend, #749028), PDCA1 BUV661 (1:200, clone “927”, BD Biosciences, #750635), CD103 BUV737 (1:200, clone “2E7”, BD Biosciences, #749393) and CD45 BUV 805 (1:800, clone “30F11”, BD Biosciences, #748370). Tumor tissue processing for 3DIP imaging Tumors were harvested and immediately fixed in BD Cytofix (diluted 1:4 in DPBS, BD Biosciences, 554655) for 20 hours at 4°C. Following fixation, tumors were washed twice with washing buffer containing 0.05% sodium azide in PBS (Sigma, #S2002-5G) and stored at 4°C in the same buffer. For immunostaining, fixed tumors were embedded in 4% low-melting agarose (Sigma-Aldrich, #A0169-250G) and sectioned into 70 um thick tumor sections using a vibratome (Leica VT1200S) equipped with conventional razor blades. Tumor sections were then washed for 15 min at room temperature on a shaker in washing buffer (dH 2 O supplemented with 0.1% Tris (Thermo Fisher, #AM9851), 1% mouse serum (Jackson ImmunoResearch, #015-000-120) and 1% ssDNA (Thermo Fisher, #15632011). Sections were then stained with 300 µl antibody staining mixture per section overnight at room temperature in the dark. The staining mixture was prepared in dH 2 O containing 0.1% Tris, 0.3% Triton-X100 (Bio-Rad, #161-0407), 1% mouse serum and 1% ssDNA to enhance permeabilization and reduce background. Following overnight incubation, tumor sections were washed twice on the shaker (10 min per wash). Finally, stained sections were mounted onto glass slides (Epredia, #J1800AMNZ) using an ImmEdge Pen (Vector Laboratories, #H-4000) and Fluoromount-G mounting medium (Thermo Fisher, #00-4958-02), then coverslipped (Roth, #BB02400500A113MNZ0) for confocal imaging. The following antibodies were used for 3DIP imaging: CD11c BV480 (1:200, clone “N418”, BD Biosciences, #746392), CD31 BV421 (1:800, clone “MEC13.3”, BioLegend, #102514), CD8 BV480 (1:200, clone “53-6.7”, BD Biosciences, #566169), CD4 BV711 (1:300, clone “RM4-5”, BD Biosciences, #563726), Ki67 AF532 (1:1200, clone “SolA15”, Thermo Fisher, #58-5698-82), CCR7 AF488 (1:200, Abcam, # Ab311015 ), MHCII AF700 (1:400, clone “I-A/I-E”, BioLegend, #107622), TCF1 PE (1:200, clone “C63D9”, CST, #6709S), PD-1 (1:300, clone “RPM1-30”, BioLegend, #109118), aGFP AF488 (1:300, Thermo Fisher, #A21311), B220 BV711 (1:200, clone “RA3-6B2”, BioLegend, #103255) and Tenascin C (1:600, clone “4C8MS”, Novus, NB110-68136AF750). Unconjugated Clec9A (1:300, R&D, #AF6776) and CD103 (1:800, R&D, #AF1990) antibodies were labeled using CF594 (Biotium, #92257) and CF633 (Biotium, #92257) conjugation kits, respectively. Unconjugated CD69 (1:200, clone “EPR25398-81”, Abcam, #ab307081) was detected using a secondary antibody, donkey anti-rabbit IgG AF555 (1:400, clone “Poly4064”, BioLegend, #406412). 3DIP imaging and image post-processing with Imaris Images were acquired using a Stellaris 5 inverted confocal microscope (Leica Microsystems) equipped with a 40× oil immersion objective with a numerical aperture of 1.3. The confocal pinhole was set to 2.00 AU, resulting in an optical section thickness of 1.038 µm. Images were captured at a resolution of 512×512 pixels and 400 Hz, with a z-spacing of 2 µm. Tiled image stacks were acquired and fluorophore spillover was calculated and subtracted using Leica LAS X software. Image segmentation was performed using Imaris software (9.7.2. and 10.1.1; Bitplane). Specific cell subsets were identified and segmented based on the following parameters: CD103: surface grain size of 0.8 µm, background subtraction: 5.69 µm, split touching objects: enabled (seed points diameter: 15 µm). MHCII: surface grain size of 1.2 µm, background subtraction: 5.69 µm, split touching objects: enabled (seed points diameter: 12 µm). CD4 + T cell: surface grain size of 1.52 µm, split touching objects: enabled (seed points diameter: 6 µm). CD8 + T cell: surface grain size of 1.52 µm, split touching objects: enabled (seed points diameter: 5 µm). For CD4 + and CD8 + T cell segmentation, the in-built machine learning segmentation module from Imaris was employed. The classifier was trained by manually labeling background and foreground (actual signal) on a 2 µm “slicer section”. The training process involved iterative labeling until optimal segmentation accuracy was achieved. For all generated cell surfaces, a minimum threshold of 150 μm 3 voxel count was applied to exclude non-specific signals and debris. Additional filtering steps and manual thresholds on the recorded fluorescent signal were applied to identify specific cell subsets of interest. These filters used the mean intensity of the respective fluorescence channels for cell type classification. To identify intracellular markers such as Ki67, TCF1 and Nur77-GFP, the Intensity Standard Deviation (“Intensity Stdv”) was used instead of the Mean. This approach was selected as the cell identifying surface was generated based on membrane markers and a high SD in fluorescence intensity within these surfaces reliably indicates positivity for a nuclear marker. To identify the cDC1 subset, CD103 + surfaces were first generated. An “Intensity Mean Clec9A” filter was then manually applied to this population, as coexpression of these markers identifies cDC1s within the TME. For quantitative analysis, statistics for each cell subset were exported to Microsoft Excel and further analyzed using custom MATLAB scripts (MathWorks) developed in-house. Results were further analyzed and plotted in GraphPad Prism software V.10.3.1. Cellular neighborhood analysis Cellular neighborhood (CN) analysis was performed using custom MATLAB scripts (MATLAB V.R2022a, using the MATLAB Compiler and Bioinformatics Toolbox). This approach involved raster-scanning tumor sections from all treatment groups with a 100 µm radius=one neighborhood. For each neighborhood, the software identified the presence and absence of the cell types of interest based on their x, y, and z coordinates. Subsequently, the neighborhoods were clustered into 10 distinct CN types using k-means clustering. The frequency of the identified CNs per sample was normalized to tumor volume to determine their prevalence across treatment groups. Results, including heatmap visualizations, were plotted using GraphPad Prism V.10.3.1. Using another custom MATLAB script, the spatial relationships between CNs, the centroid of each individual neighborhood was detected and the absolute number of direct connections between a given neighborhood and other neighborhoods was quantified and visualized. It is important to note that it is not a fully custom tool but rather a user interface designed to enable fit-for-purpose usage of established and published algorithms (eg, Cytomap 31 or similar tools, 32 standard functions and simplified implementations of available algorithms in the Roche IT environment. Automated detection and quantification of T cell-DC clusters To evaluate the prevalence of intratumoral aggregates characterized by CD4 + and CD8 + T cell clusters colocalizing with cDC1s (=TDCs), an automated digital image analysis workflow was developed using HALO image analysis software (V.3.6.4134.396, Indica Labs). Following an initial reduction to two-dimensional maximal intensity projections, a single DenseNet V2 convolutional neural network classifier was trained to identify T cell-rich cellular clusters accurately. Manually annotated regions used for training came from all experimental treatment groups and constituted visually identified TDCs of varying sizes, characterized by dense CD4 + and CD8 + T cell aggregation. The model underwent deep learning optimization until convergence, defined as a cross-entropy loss value <0.1. Subsequently, the trained classifier was applied to analyze the full image set, and an iterative refinement involving cycles of annotation, deep learning model training, visual expert validation, and subsequent correction was performed. This approach continued until the incidence of misclassified regions was minimized and no deviations were observed across further iterations prior to generating tissue masks. Next, an Area Quantification algorithm (V.2.4.3) was parameterized to characterize T cell aggregates colocalizing with cDC1s (=TDCs). This analysis module screened classified T cell-rich clusters based on the coexpression of CD103 and Clec9A markers. The algorithm worked by first identifying individual cells within the defined T cell cluster masks and only clusters with constituent DCs exhibiting coexpression of both markers above a set intensity threshold were retained and counted as true TDCs. These rigorous criteria ensured the unambiguous identification of aggregates containing both T cells and cDC1s. No specific diameter threshold was pre-defined for TDC identification. The number of TDCs identified in each treatment group was then normalized to tumor tissue volume and visualized using GraphPad Prism software 10.3.1. Dissociation of tumor samples Tumor samples were harvested into cold PBS and stored on ice until processing. Tumors were transferred to GentleMACS C Tubes (Miltenyi Biotec, #103-093-237) containing 2.5 mL of enzymatic digestion mixture consisting of RPMI 1640, 2% FBS and 0.05 mg/mL DNase I (Sigma-Aldrich, #10104159001) and 0.25 mg/mL Liberase TL (Roche, #5401020001). Tumors were minced into small pieces using scissors. The tubes were then placed on a GentleMACS Octo Dissociator (Miltenyi Biotec) using the manufacturer’s recommended tumor dissociation protocol, followed by incubation for 30 min at 37°C. Following incubation, the cell suspension was filtered through 70 µm MACS SmartStrainers (Miltenyi Biotec, #130-110-96) and washed with RPMI 1640 containing 2% FBS. The samples were centrifuged and resuspended in pZerve freezing medium (Sigma-Aldrich, #Z1653) and stored for further downstream analysis, including ex vivo restimulation assays and single-cell RNA sequencing (scRNA-seq) analysis. Ex vivo restimulation and intracellular cytokine staining Frozen tumor single-cell suspensions were thawed, resuspended and plated at 1 Mio. cells/ml in RPMI 1640 medium supplemented with 0.1% 2-Mercaptoethanol (Gibco, #31350-010) in U-bottom 96-well plates. Cells were stimulated with 50 ng/mL phorbol 12-myristate 13-acetate (PMA, Sigma Aldrich, P1585) and 1 µM Ionomycin (Sigma Aldrich, I0634) for 4 hours at 37°C. After 1 hour of stimulation, Golgistop (1:1500, BD Biosciences, #554724) and Golgiplug (1:1000, BD Biosciences, 555029) were added to block cytokine release. The PMA restimulation mix and Golgistop/Golgiplug were prepared in RPMI 1640 medium with 0.1% 2-Mercaptoethanol. Following stimulation, cells were washed with PBS and incubated with CD16/CD32 Fc receptor blocking antibody (1:100, BD Biosciences, #553142) and Fixable Blue viability dye (1:400, Thermo Fisher, # L34962 ) for 20 min at 4°C. After a washing step, cell surface staining was added to the cells for 20 min at 4°C. After two washing steps, cells were fixed and permeabilized using eBioscience Fixation/Permeabilization buffer (Thermo Fisher, #88-8824-00) for 20 min. at room temperature. Following fixation, cells were washed twice with eBioscience permeabilization buffer (1:10 in H2O, Thermo Fisher, #00-8333-56). An intracellular staining mixture, prepared in eBioscience permeabilization buffer, was added to the cells. Cells were incubated overnight at 4°C, followed by two washes and filtration prior to flow cytometric analysis. Data were acquired on a BD LSRFortessa flow cytometer and analyzed using FlowJo software (V.10.10.0) The following antibodies were used for cell surface staining following restimulation: CD49b ef450 (1:200, clone “DX5”, Thermo Fisher, #48-5971-82), CD19 BV605 (1:200, clone “1D3”, BD Biosciences, #563148), CD8 BV711 (1:200, clone “53-6.7”, BioLegend, #100748), CD4 PE (1:400, clone “H129.19”, BioLegend, #130319), CD45 AF700 (1:200, clone “30-F11”, Thermo Fisher, #56-0451-82) and T-cell receptor beta (TCRβ) BV510 (1:100, clone “H57-597”, BD Biosciences, #563221). The following antibodies were used for intracellular staining following restimulation: TCRβ BV510 (1:100, clone “H57-597”, BD Biosciences, #563221), interferon-gamma (IFN-γ) APC (1:150, clone “XMG1.2”, Thermo Fisher, #17-7311-82), IL2 PerCP Cy5.5 (1:200, clone “JES6-5H4”, BioLegend, #503822) and tumor necrosis factor-alpha (TNFα) PE Cy7 (1:200, clone “MP6-XT22”, BioLegend, #506324). Sample preparation for single-cell RNA-sequencing Frozen tumor single-cell suspensions were thawed and washed with RPMI 1640 supplemented with 10% FBS. Following centrifugation, cells were washed once with PBS. To block non-specific binding, normal Mouse IgG (1:50 dilution in FACS buffer, Thermo Fisher, #10400C) was added to the cells and incubated for 10 min at 4°C. After one PBS wash, cells were resuspended with the antibody staining mix and incubated for 30 min at 4°C. Following two additional washes, cells were resuspended in FACS buffer. Cell sorting was performed using a 100 µm nozzle with cells separated into two distinct populations: CD45 + CD3 + NK1.1 + cells (T and natural killer (NK) cells) and CD45 + CD3 − NK1.1 − (myeloid cells). This sorting strategy was employed to enrich for the rare cDC1 subset within the myeloid population. Sorted cells were collected in PBS containing 1% bovine serum albumin (Sigma-Aldrich, #126615-25ML). For each sample, myeloid cells and NK/T cells were mixed at a ratio of 5:1 (myeloid cells:T/NK cells) immediately prior to processing for scRNA-seq. The following antibodies were used for cell sorting: CD45 AF700 (1:100, clone “30F11”, Thermo Fisher, #56-0451-82), CD3 PE CF594 (1:100, clone “145-2C11”, BD Biosciences, #562286), NK1.1d PE (1:400, clone “PK136”, eBioscience, #12-5941-82) and live/dead NIR (1:400, Thermo Fisher, # L34976 ). The following additional antibodies were used to gate (including and excluding markers) and identify DCs within the sorted samples: MHCII ef450 (1:400, clone “114.15.2”, Thermo Fisher, #14-5321-82), CD11c APC (1:200, clone “N418”, BioLegend, #117310), B220 FITC (1:100, clone “RA3-6B2”, BioLegend, 103206), Ly6C BV605 (1:100, “clone HK1.4”, BioLegend, #128036), F4/80 PeCy5 (1:100, clone “BM8”, BioLegend, #123112). scRNAseq of tumor samples Tumor samples were processed according to the protocols described in “Dissociation of tumor samples” and “Sample preparation for single-cell RNA sequencing” sections. A total of 16 samples were analyzed (from four treatment groups: vehicle, FAP-CD40, PD1-IL2v and FAP-CD40+PD1-IL2v), with four mice per group. The cell number and viability of sorted and pooled cells were determined using a Nexcelom Cellometer Auto 2000. A total of approximately 18,000 viable cells per sample were loaded into the 10x Genomics Chromium Connect automated platform and library preparation was performed according to the manufacturer’s indications (Chromium Next GEM Automated Single Cell 5’ Reagent Kit V.2). 14 and 16 PCR cycles were used for complementary DNA (cDNA) and library amplification respectively. The concentration of the cDNAs and the resulting final libraries were quantified using an Invitrogen Qubit fluorometer (High Sensitivity dsDNA Assay) and the respective fragment size distribution was determined using an Agilent 2100 Bioanalyzer automated electrophoresis instrument. The libraries were pooled equimolecularly and sequenced on a S2 flow cell of a NovaSeq 6000 system (Illumina). Sequencing followed the 10x Genomics read configuration (R1=26, i7=10, i5=10, R2=90) and achieved a depth of approximately 15,000 reads per cell. Computational analysis of scRNAseq samples Raw sequencing reads were mapped to the GRCm38.84 genome via Cell Ranger Single Cell software V.7.1.0 (10x Genomics) using the prebuilt reference refdata-gex-mm10-2020-A provided by 10x Genomics. Per-sample raw gene expression matrices were merged and analyzed using the besca package. 33 Low-quality cells and potential multiplets were filtered out based on defined thresholds (minimum 400 genes, 800 counts; maximum 5,000 genes, 30,000 counts; no more than 5% mitochondrial genes). This filtering step retained 57,320 cells in total for analysis, with 2,000 to 6,000 cells per sample. Normalized gene expression was obtained by log-transforming Unique Molecular Identifier (UMI) counts per 10,000 reads [log(CP10K+1)]. After scaling, the most variable genes were determined using specific mean and dispersion thresholds. A subset of 2,412 highly variable genes, present in at least 50% of samples, was selected for principal component analysis. The first 50 principal components were used to calculate the 10 nearest neighbors, and the resulting neighborhood graph was embedded into a two-dimensional space via a Uniform Manifold Approximation and Projection (UMAP). 34 Clustering of cells was performed using the Leiden algorithm. 35 Cell type annotation Cell annotation was conducted using the semi-automated Sig-annot module within the besca package. The gene sets employed for different cell types can be accessed at https://github.com/bedapub/besca/blob/main/besca/datasets/genesets/CellNames_scseqCMs6_sigs.gmt . Clusters were initially assigned cell types at varying levels of granularity. Subsequently, manual inspection was performed to resolve cluster mixtures, particularly for DC types. Further refinement of cell type annotations was achieved by applying heuristic cut-offs to signature scores within specific clusters to reannotate individual cells. Per-cell signature scores were calculated using the scanpy.tl.score_genes function with default parameters and besca signatures. Doublet detection was performed using the R package scDblFinder (V.1.14.0) 36 in R V.4.3.0; cells identified as doublets were excluded from all downstream analyses. Visualizations, such as the expression levels of selected genes across cell types, were generated using custom scripts primarily based on besca and scanpy functions. Differential gene expression analysis For each cell type, cells were summed into pseudobulks by sample, and pseudobulk samples with fewer than 10 cells were excluded from further analysis. The analysis followed the Orchestrating Single-Cell Analysis with Bioconductor guidelines using edgeR (V.3.42.2) 37 38 in R V.4.3.0. To extract the treatment effect, each treatment was compared with the control vehicle treatment. To remove unwanted technical variation, the RUVSeq R package (V.1.34.0) 39 was used to compute factors of unwanted variation (Wk). Based on the examination of p value histograms, W_1 and W_2 were included or not as covariates in the models to remove unwanted variation. To assess any synergistic effect present only in the combined treatment, the combined treatment was compared individually to the monotreatments using the same procedure. Gene set enrichment analysis Gene set enrichment analysis (GSEA) was performed to identify significantly enriched pathways from differential gene expression (DGE) results. A ranked gene list was generated based on the log2 fold change values obtained from the DGE analysis. GSEA was implemented in R V.4.3.0 using the metabaser package (V.4.9.3), specifically employing the GSEA.preranked function with gene sets provided by the CBDD package (V.17.3.0) to identify enriched biological pathways. The visualizations of pathways and genes assigned to pathways were generated using ggplot2 (V.3.5.1) and scDotPlot (V.1.1.0). T cell gene signature scoring To investigate the impact of treatment on T cell states and function, gene signature scores reflecting core T cell functions were calculated for each T cell subset independently. Signatures, sourced from selected publications, 40 41 were filtered to retain only genes expressed in a minimum of 5% of the cells. Adjusted neighborhood scoring 42 was subsequently employed to derive per-sample signature scores. Statistical significance of differences in scores between groups was evaluated using a one-way analysis of variance and Benjamini and Hochberg procedure for multiple testing correction. Statistical analysis Statistical significance was evaluated using GraphPad Prism V.10.3.1. The statistical tests applied for each analysis are indicated in the respective figure legends. Results FAP-CD40 treatment remodels the pancreatic tumor microenvironment through cDC1 activation and T cell recruitment Given the critical role of cDC1s in initiating antitumor immunity, we first sought to evaluate the effects of FAP-CD40 monotherapy on cDC1 activation and remodeling of the immune-cold pancreatic TME. Additionally, we aimed to assess whether FAP-CD40 treatment could enhance infiltration of T cells, including potential antigen-specific CD8 + T cells. To this end, we treated immunocompetent syngeneic C57BL/6 mice bearing subcutaneous KPC tumors with FAP-CD40. A single-dose administration of FAP-CD40 on day 14, when tumors reached 200 mm 3 , significantly delayed tumor growth compared with untreated tumors ( figure 1A ). We analyzed tumor-infiltrating cDC1s (identified by coexpression of Clec9A, CD103, MHCII and CD11c expression; online supplemental figure 1A ) via 3DIP imaging and detected morphological changes of cDC1s induced by FAP-CD40 24 hours post-treatment ( figure 1B ). While cDC1s in untreated tumors typically displayed a round morphology, those in treated tumors exhibited an elongated shape in specific regions, characteristic of activated and migratory DCs. 43 We further performed sphericity analysis at 24 hours post-treatment, which revealed heterogeneous cDC1 shapes, ranging from round to highly elongated with extended cellular projections ( figure 1C ). Sphericity values approaching 1 indicate a perfect sphere, while lower values represent less spherical structures. 44 Based on literature and qualitative assessment of the dataset shown in figure 1C , a sphericity index below 0.6 was used to classify cells as more elongated. Quantification showed a shift toward more elongated morphologies following FAP-CD40 (n=2–3), with 51% of cDC1s displaying a sphericity<0.60 compared with 36% in controls ( figure 1D ), suggesting treatment-associated morphological remodeling. Figure 1. FAP-CD40 monotherapy delays pancreatic KPC tumor growth and drives changes in cDC1 morphology and spatial distribution. ( A ) Tumor growth kinetics comparing untreated (vehicle) with FAP-CD40-treated mice (n=15 mice/group at study start). Statistical analysis (unpaired t-test with Welch’s correction) was performed on mice alive on day 24 (vehicle: n=10; FAP-CD40: n=9; *p≤0.05). ( B ) Representative confocal images showing morphological changes in cDC1s (expressing CD103 and Clec9A) within KPC tumors 24 hours after FAP-CD40 treatment compared with untreated controls (scale bars: 50 µm). ( C ) Representative images illustrating the spectrum of cDC1 morphologies and corresponding sphericity values within a single FAP-CD40-treated tumor section. Frequency distribution analysis of cDC1 sphericity was performed on cDC1 populations from two to three individual tumors/group at 24 hours post-treatment. A sphericity threshold of 0.60 was used to distinguish more spherical (>0.60) from elongated (<0.60) cDC1s. FAP-CD40-treated tumors show a higher frequency of cDC1s with sphericity <0.60. Statistical significance was assessed using an unpaired t-test with Welch’s correction (****p≤0.0001). ( D ) Frequency of elongated cDC1s (sphericity<0.60) across treatment groups (n=2–3 individual tumors/treatment group). ( E ) Quantitative analysis of cDC1s in KPC tumors following FAP-CD40 treatment or control at 72 hours and 8 days post-treatment, assessed by 3DIP imaging (n=3 individual tumors/treatment group). Statistical significance was determined using an unpaired t-test with Welch’s correction within each time point (*p≤0.05). Data are presented as mean±SD. ( F ) Spatial distribution of cDC1s in untreated and FAP-CD40-treated KPC tumors at 72 hours and 8 days post-treatment. Representative images show one tumor section per group and time point (scale bars: 1 mm). Black dashed lines indicate the tumor border. cDC1s, conventional type 1 dendritic cells; 3DIP, 3D immune phenotyping; FAP, fibroblast activation protein. Open in a new tab To further characterize the effect of FAP-CD40 on cDC1s, we analyzed their abundance and spatial distribution following treatment ( figure 1E and F ). Intratumoral cDC1 numbers showed a decreasing trend within 72 hours of treatment and remained significantly reduced through day 8 compared with untreated controls. While our spatial analysis confirmed reduced DC density within treated tumors, we observed a corresponding 2.5-fold increase in cDC1 counts within tdLNs at 72 hours post FAP-CD40-treatment ( online supplemental figure 1B ). We further examined the spatial distribution of cDC1s within tdLNs at this time point and found enhanced accumulation of these cDC1s following FAP-CD40 treatment, potentially within the subcapsular sinus region ( online supplemental figure 1C ). These observations, consistent with previous studies using untargeted CD40 agonists, 12 indicate that FAP-CD40 treatment induces cDC1 activation and subsequent migration from the tumor to dLNs, enabling tumor antigen presentation and priming of CD8 + T cells within secondary lymphoid organs. Combination of FAP-CD40 with PD1-IL2v treatment promotes the formation of T cell-dendritic cell clusters and enhances antitumor response While FAP-CD40 monotherapy delayed tumor growth, it failed to induce effective tumor regression in KPC tumors. To improve therapeutic efficacy, we investigated the effects of combining FAP-CD40 with PD1-IL2v, known for its ability to activate, expand and differentiate intratumoral T cells. In this combinatorial approach, we treated mice bearing tumors of approx. 200 mm 3 with a single dose of FAP-CD40 on day 13, simultaneously with the first dose of PD1-IL2v and continued PD1-IL2v treatment with weekly administration over 3 consecutive weeks ( figure 2A ). The combination of FAP-CD40 with PD1-IL2v resulted in significant tumor regression, while monotherapies alone led to only modest tumor growth inhibition ( figure 2B and online supplemental figure 2A ). This demonstrates a potent synergy between the two therapeutic agents. Figure 2. FAP-CD40+PD1-IL2v combo therapy induces tumor regression and drives CD4+/CD8+T cell infiltration 10 days post-treatment. ( A ) Timeline of therapy administration (post-treatment days). FAP-CD40 was administered at a single dose on day 0 (tumor volume~200 mm 3 ). PD1-IL2v was given weekly on days 0, 7 and 14. ( B ) Tumor growth curves showing the average tumor volume across treatment and control groups throughout the study (data pooled from four individual experiments with n=12–15 mice/group per experiment). Statistical analysis was performed on measurements taken on day 21 post-treatment start. Error bars represent SEM. Statistical significance was determined by one-way ANOVA with Tukey’s post hoc test (*p≤0.05, **p≤0.01, ****p≤0.0001). ( C ) Quantitative analysis of cDC1s (CD103 + Clec9A + ) and mregDCs (MHCII + CCR7 + ) within KPC tumors across treatment and control groups (n=3–4 individual tumors/treatment group). ( D, E ) Quantitative analysis of CD4 + , CD8 + T cells and their corresponding PD-1 + TCF1 + subsets within KPC tumors across treatment and control groups (n=3–4 individual tumors/treatment group). ( F ) Spatial distribution of CD4 + and CD8 + T cells within KPC tumors across treatment and control groups. Representative images show one tumor section from each treatment group (scale bars: 1 mm). Black dashed lines indicate the tumor border. For panels ( C–E ), data are presented as median (IQR). Statistical significance was determined using the Kruskal-Wallis-test followed by Dunn’s post hoc test, with targeted comparisons to the combo group (*p≤0.05, **p≤0.01). ANOVA, analysis of variance; cDC1s, conventional type 1 dendritic cells; FAP, fibroblast activation protein; IL2v, interleukin 2v; mregDCs, mature dendritic cells enriched in immunoregulatory molecules; PD1, programmed cell death protein-1. Open in a new tab To further investigate immune remodeling induced by the combo treatment, we performed 3DIP analysis of tumor tissues 10 days post-treatment. Details of the multiplex imaging panel and representative images of cell segmentation for CD4 + and CD8 + T cells, cDC1s and mregDCs (mature dendritic cells enriched in immunoregulatory molecules) are provided in online supplemental figure 2B and C . As depicted in figure 2C , quantitative assessment of cDC1s showed comparable cell frequencies between untreated and PD1-IL2v-treated groups. Consistent with previous findings, FAP-CD40 monotherapy was associated with a reduction in intratumoral cDC1s. When combined with PD1-IL2v, there was a trend toward increased cDC1 numbers, suggesting a potential reversal of this decrease. However, this difference was not statistically significant ( figure 2C ). In contrast, frequencies of mregDCs (characterized by MHCII and CCR7 expression but lacking the classical cDC1 markers CD103 and Clec9A; online supplemental figure 3A ) remained unchanged across all treatment groups in comparison to untreated mice. Furthermore, we found distinct localization patterns of cDC1s and mregDCs in untreated tumors ( online supplemental figure 3B ). Following FAP-CD40 or PD1-IL2v treatment, whether as monotherapy or in combination, both DC subsets colocalized within the same regions of the tumor tissues (at the tumor edge and within the tumor parenchyma), suggesting their coordinated migration to areas of inflammation to facilitate antitumor responses ( online supplemental figure 3B ). Moreover, while FAP-CD40 monotherapy showed only a modest trend toward increased CD4 + T cell numbers, the combination of FAP-CD40 with PD1-IL2v induced a substantial 20-fold expansion of CD4 + T cells compared with untreated controls ( figure 2D ). Combination therapy increased not only the overall CD4 + T cell population but also multiple CD4 + T cell subsets in parallel, including TCF1 + PD-1 + CD4 + T cells, known for its stem-like properties and effector differentiation potential, 45 as well as the proliferating Ki67 + CD4 + T cells ( figure 2D and online supplemental figure 3C ). CD8 + T cell populations exhibited a similar pattern, with increases in total CD8 + T cells accompanied by parallel expansion of both the stem-like subset (TCF1 + PD-1 + ) 23 46 47 and Ki67 + subset upon combo treatment, although to a lesser extent compared with CD4 + T cells ( figure 2E and online supplemental figure 3D ). Beyond changes in T cell numbers, differences in intratumoral T cell distribution were observed across treatment groups. While CD4 + and CD8 + T cells were primarily confined to the tumor edge in untreated and PD1-IL2v-treated tumors, T cells were more frequently observed within the tumor core in FAP-CD40-treated tumors, with a more pronounced intratumoral presence in the combo-treated group ( figure 2F ). A detailed spatial analysis following the different treatment regimens revealed the emergence of organized clusters of T cells and DCs (TDCs) within the TME, mainly in the combo therapy group, with only sparse TDCs detected in monotherapy and control groups. These TDCs consisted of densely packed clusters of CD4 + and CD8 + T cells colocalizing with cDC1s ( figure 3A,B ) and were distributed throughout the combo-treated tumor tissue ( figure 3A ), while containing relatively few mregDCs ( online supplemental figure 4A and B ). Interestingly, the organization of CD4 + T cells in combo-treated tumors, particularly within these TDCs, resembled the characteristic dense arrangement of CD4 + and CD8 + T cells found in T cell zones of LNs ( online supplemental figure 4C and D ). In contrast, intratumoral CD4 + T cells in untreated or monotherapy groups appeared sparse and less interconnected. Figure 3. FAP-CD40+PD1-IL2v combo therapy induces formation of intratumoral TDCs 10 days post-treatment, which are associated with improved therapeutic response. ( A ) Spatial characterization of TDCs within the tumor tissue across treatment groups. Upper panel: yellow circles denote individual CD4 + /CD8 + T cell clusters, whereas yellow circles outlined by a white rectangle indicate TDCs (CD4 + /CD8 + T cell clusters filtered for cDC1 presence), identified using an automated digital image analysis workflow in HALO (scale bars: 1 mm). The middle panel shows a magnified view of CD4 + and CD8 + expression within a representative TDC (red rectangle; scale bars: 50 µm). The lower panel highlights the presence of CD103 + Clec9A + cDC1s within the same TDC (scale bars: 50 µm). ( B ) High-magnification images of TDCs, showing aggregation of CD4 + and CD8 + T cells colocalizing with cDC1s across treatment groups (scale bars: 20 µm). ( C ) Quantitative analysis of TDCs across treatment groups. Data represent pooled results from three independent experiments (n=9–12 individual tumors/treatment group). Statistical significance was determined by one-way ANOVA with Dunnett’s post hoc test, with targeted comparisons to the combo group (***p≤0.05, ****p≤0.0001). ( D ) Quantification of TDCs in treatment-responding (“Resp.”, n=15 tumors) versus non-responding (“Non-R.”, n=25 tumors) tumors across treatment groups. Response was defined as tumor regression occurring between days 15 and 20 prior to tumor tissue harvesting. This analysis demonstrates the relation between TDC formation and therapeutic outcome. Statistical significance was evaluated using an unpaired t-test with Welch’s correction (**p≤0.01). ANOVA, analysis of variance; cDC1, conventional type 1 dendritic cell; FAP, fibroblast activation protein; IL2v, interleukin 2v; PD1, programmed cell death protein-1; TDCs, T cell-DC clusters. Open in a new tab Quantitative image analysis further established TDCs as a defining feature of combo therapy, with significantly higher numbers observed in tumors receiving the combo treatment compared with monotherapy or vehicle control groups ( figure 3C ). Moreover, responding tumors consistently displayed high TDC densities (>19 TDCs/mm 3 ), while the non-responding ones showed minimal TDC formation (mean of 3 TDCs/mm 3 ), as shown in figure 3D . Importantly, TDCs were distinct from tertiary lymphoid structures (TLSs), as they lack a prominent B cell component ( online supplemental figure 4E ). In summary, our spatial analysis identified TDCs as a key feature of combination therapy with FAP-CD40 and PD1-IL2v, occurring at higher numbers, widely distributed across the tumor, and associated with improved treatment response. Intratumoral TDCs drive therapeutic response independent of LN priming post-treatment To investigate whether the antitumor response following combo therapy depended on de novo T cell priming in the LN or was driven by pre-existing, primed T cells within the tumor, we used FTY720 to block T cell egress from LNs ( figure 4A ). FTY720 acts by inducing the internalization of the sphingosine 1-phosphate 1 receptor on lymphocytes, thereby blocking the exit of T cells from the thymus and LNs. 48 We administered FTY720 2 days prior to therapy start and continued FTY720 administration every 4 days until study termination. Peripheral blood analysis by flow cytometry confirmed effective T cell depletion 1 day before treatment initiation, with sustained depletion observed through day 35, 4 days prior to study endpoint ( figure 4B ). Notably, combo therapy led to regression of KPC tumors even with FTY720-mediated blockade of T cell egress from the LNs ( figure 4C ) and still elicited the highest intratumoral infiltration of CD4 + and CD8 + T cells, including their Ki67 + and TCF1 + PD-1 + subsets, among all groups ( figure 4D and online supplemental figure 5A and B ). This suggests that therapeutic efficacy is largely driven by local proliferation and activation of tumor-resident T cells, rather than requiring continuous recruitment of newly primed T cells from lymphoid organs. Figure 4. Efficacy of FAP-CD40+PD1-IL2v combo therapy functions independently of LN priming, with comparable immune cell compositions observed between FTY720-treated and untreated combo treatment groups. ( A ) Experimental design schema illustrating LN egress inhibition of T cells using FTY720, preventing T cell trafficking from LNs to peripheral tissues. ( B ) Quantification of T cells in peripheral blood 1 day before start of therapy and 4d prior to termination, comparing FTY720-treated and untreated groups (n=4 mice/group). Statistical analysis was performed using an unpaired t-test (*p≤0.05, ****p≤0.0001). ( C ) Tumor growth curves showing the average tumor volume across treatment groups over time (n=14 mice/group at study start), including an additional group receiving combo therapy+FTY720 treatment. Statistical analysis (one-way ANOVA with Tukey’s post hoc test) was performed on measurements taken on day 36 post tumor inoculation (*p≤0.05, **p≤0.01, ***p≤0.001, ****p≤0.0001). Error bars represent SD of the mean. Data shown are representative of a single experiment. ( D ) Quantitative analysis of CD4+ and CD8+ T cells and their respective functional subsets within KPC tumors across treatment and control groups±additional FTY720 treatment. Cell populations were quantified using 3DIP imaging. Statistical significance was determined using the Kruskal-Wallis test followed by Dunn’s post hoc test, with targeted comparisons to the combo group. Asterisks indicate significance (*p≤0.05, **p≤0.01). Only the comparison between the combo group and vehicle control reached statistical significance, all other comparisons were not statistically significant (p>0.05). ( E–F ) Raster scanning analysis of tumor tissues across treatment groups without ( E ) or with ( F ) additional FTY720 treatment, revealing 10 distinct CNs. Heatmap indicates the presence or absence of specific cell subsets within each CN. ( G ) Quantification of CNs showing significantly enriched CNs (CN1 and CN7) in combo-treated tumor tissues. Data points represent pooled samples from both FTY720-treated and FTY720-untreated tumors within each treatment group (n=6–8 individual tumors/treatment group). ( H ) Quantification of CNs showing enriched CNs (CN6 and CN10) in tumors treated with the less efficacious FAP-CD40 and PD1-IL2v monotherapies or control. Data points represent pooled samples from FTY720-treated and FTY720-untreated tumors within each treatment group (n=6–8 individual tumors/treatment group). Statistical significance in ( G–H ) was determined by one-way ANOVA with Dunnett’s post hoc test, with targeted comparisons to the combo group. Data presented as mean±SD (*p≤0.05, **p≤0.01, ****p≤0.0001). ANOVA, analysis of variance; cDC1s, conventional type 1 dendritic cells; DC1, type 1 dendritic cell; 3DIP, 3D immune phenotyping; FAP, fibroblast activation protein; IFNγ, interferon-gamma; IL2v, interleukin 2v; LN, lymph node; PD1, programmed cell death protein-1; TNFα, tumor necrosis factor-alpha. Open in a new tab To comprehensively investigate the relationship between tissue architecture and treatment response, we performed 3DIP imaging followed by systematic raster-scanning of cell types across entire tumor sections from all treatment groups collected 10 days after the first treatment ( figure 4E,F ). This unbiased, whole-tumor section approach enabled a detailed mapping of CNs and identified spatial relationships among various immune and vascular cell types. Within each region (at a radius of 100 µm), we quantified the presence of cDC1s, CD4 + T cells, CD8 + T cells, their respective functional subsets, mregDCs and blood vessels. Computational analysis of the TME identified ten distinct CNs within the analyzed tumor tissues ( figure 4E and online supplemental figure 6A ), including a specific neighborhood (CN1) corresponding to the previously qualitatively described TDCs, which are enriched with cDC1s, CD4 + and CD8 + T cells ( figure 4E and G ). An equivalent analysis of tumors from mice treated with FTY720 showed a similar tissue architecture and CNs to the FTY720-untreated controls, indicating that FTY720 blockade did not significantly alter overall spatial organization of the TME ( figure 4F ). Moreover, the CN1 representing TDCs was maintained despite blockade of T cell egress, confirming that the TDCs do not depend on recruitment of LN-primed T cells ( figure 4F and online supplemental figure 6B ). Combination therapy also increased the presence of CN7 ( figure 4G ), which exhibited similar CD4 + and CD8 + T cell infiltration as CN1 but lacked cDC1s. Proximity analysis further demonstrated strong spatial correlation between these two CNs, suggesting a connected microenvironment in which DCs may migrate between CNs to facilitate T cell activation ( online supplemental figure 6C ). In contrast, the less efficacious monotherapy groups FAP-CD40 and PD1-IL2v showed enrichment of CN6 and CN10 ( figure 4H ), characterized by cDC1 and CD8 + T cell colocalization but minimal CD4 + T cell presence, or cDC1s alone, respectively ( figure 4H ). Representative spatial localization of these CNs within the tumor tissue, including high-resolution views of CN1, CN7, CN6 and CN10, is shown in online supplemental figure 6D . Taken together, our spatial analysis revealed that distinct CNs patterns correlate with treatment efficacy. Combination therapy with FAP-CD40 and PD1-IL2v uniquely enriched CN1, the CN containing TDCs, as well as the adjacent T cell-rich CN7. Notably, both CNs persisted despite FTY720-mediated blockade of T cell egress from LN, indicating independence from continuous LN priming. In contrast, less effective treatments were dominated by CNs with cDC1 and CD8 + T cells colocalization, but reduced CD4 + T cell presence. CNs containing solely cDC1s, prevalent in untreated and monotherapy tumors, lacked the ability to recruit additional T cells. Antitumor efficacy of the combo treatment depends on both CD4 + and CD8 + T cell subsets at treatment onset To dissect the individual and combined contributions of CD4 + and CD8 + T cell subsets to the therapeutic efficacy of the combo treatment, we performed antibody-mediated T cell depletion studies. We treated mice bearing KPC tumors with depleting antibodies against CD4 + T cells, CD8 + T cells, or both, on days 6, 9 and 12 post-tumor inoculation ( figure 5A ). We initiated FAP-CD40 and PD1-IL2v combo treatment on day 13 when tumors reached approximately 200 mm 3 and continued PD1-IL2v administration weekly for a total of 3 weeks. We confirmed successful and specific depletion of the targeted T cell populations in peripheral blood by flow cytometry 1 day prior to treatment initiation ( online supplemental figure 7A and B ). Figure 5. Antitumor efficacy of FAP-CD40+PD1-IL2v combo therapy critically depends on both CD4 + and CD8 + T cell subsets present at treatment initiation. ( A ) Experimental design schema showing the time points of therapy administration throughout the experiment. FAP-CD40 was administered once on day 13 (tumor volume ∼200 mm 3 ). PD1-IL2v was given weekly on days 13, 20 and 27. Anti-CD4+and/or anti-CD8+ depletion antibodies were administered on day 6, 9 and 12. ( B ) Tumor growth curves showing the average tumor volume across treatment groups over time (n=12 mice/group at study start). Statistical analysis was performed on measurements taken on day 21 post-treatment start. Error bars (only upward shown) represent SD of the mean. Data shown are representative of two independent experiments. Statistical analysis was performed by one-way ANOVA with Tukey’s post hoc test (*p≤0.05, ***p≤0.001). ( C ) Representative confocal images of KPC tumors showing intratumoral CD4 + and CD8 + T cell distribution at treatment start across treatment groups (control vs CD4 + , CD8 + or CD4 + /CD8 + T cell-depleted mice). Scale bars: 50 µm. ANOVA, analysis of variance; FAP, fibroblast activation protein; IL2v, interleukin 2v; i.p., intraperitoneal; i.v., intravenous; PD1, programmed cell death protein-1; s.c., subcutaneous. Open in a new tab Evaluation of tumor growth inhibition revealed that both CD4 + and CD8 + T cells together are critical for the full therapeutic effect of the combo treatment ( figure 5B and online supplemental figure 7C ). Depletion of either subset resulted in only partial loss of efficacy, with modest tumor growth inhibition. In contrast, simultaneous depletion of both CD4 + and CD8 + T cells completely abrogated the antitumor response, resulting in tumor growth kinetics comparable to those of untreated controls. Further analysis of the TME by 3DIP imaging confirmed that CD4 + and/or CD8 + T cell depletion was effective at the initiation of therapy ( figure 5C ). Altogether, these findings demonstrate that the combo treatment relies not only on CD8 + T cells, which are known targets of PD1-IL2v, 21 but also on the copresence of CD4 + T cells within the TME to generate potent antitumor responses. Combination therapy promotes T cell activation and Th1 polarization To characterize the T cell phenotypes across treatment conditions and further dissect the molecular mechanisms underlying the increased efficacy observed with the combo therapy approach, we performed scRNA-seq of tumor samples from each treatment group. Tumors were harvested 10 days after the first treatment and processed for cell sorting. T/NK cells (CD45 + CD3 + NK1.1 + ) and myeloid populations (CD45 + CD3 − NK1.1 − ) were sorted separately to enrich for the rare cDC1 subset. The scRNAseq analysis identified distinct cell populations within the TME, as illustrated in online supplemental figure 8 , which provides an overview of the cell types, their marker gene expression patterns and relative abundances across treatment groups. Pathway enrichment analysis showed enhanced upregulation of IL-12 and IL-18 signaling within CD4 + T cells from combo-treated tumors compared with vehicle controls, alongside enhanced activity in CD28 costimulatory, PD-1 and TCR alpha/beta signaling pathways ( figure 6A ). Analysis of differentially expressed genes within these pathways further identified increased expression of transcripts associated with T cell activation and T helper cell 1 (Th1) differentiation ( figure 6B ), including CD40lg, Tbx21 (encoding T-bet), CD28, IL-18 signaling components (Il18ra, IL18r1), Stat4, and Il2rb—genes associated with Th1-type responses. 49 , 53 In addition to Th1-associated genes, we found an upregulation of gene signatures related to T cell adhesion and chemokine/chemokine receptor interactions ( figure 6C,D ), including elevated expression of Cxcr6, Cxcr4, Ccr5 and Itgb1, suggesting enhanced migratory and tissue-retention capacity. 54 , 58 Figure 6. Molecular and functional characterization of T cells in response to combo therapy reveals Th1 polarization, enhanced TCR signaling and TDC formation that correlates with therapeutic efficacy. ( A ) Enriched Metabase pathways within CD4 + T cells following combo treatment compared with control, identified by single-cell RNA sequencing. ( B ) Heatmap showing upregulated genes in CD4 + T cells following combo therapy compared with control, focusing on pathways identified in ( A ). ( C ) T cell phenotypes enriched in CD4 + T cells on combo treatment. FDR-adjusted p values: T-helper, 0.0010; CD4 Adhesion, 0.0013; CD4 Chemokine/Chemokine receptor, 0.0010. ( D ) Heatmap of genes upregulated in CD4 + T cells following combo therapy, focusing on genes enriched within the T cell phenotypes identified in ( C ). Data in ( A–D ) are representative of a single experiment (n=4 individual tumors/treatment group). ( E, F ) Intracellular cytokine analysis of IFN-γ and TNFα secretion by intratumoral CD4+and CD8+ T cells following ex vivo PMA/Ionomycin restimulation at study termination across treatment groups. Data shown are representative of three individual experiments. ( G ) Intracellular cytokine analysis of IFN-γ and TNFα secretion by intratumoral CD3 + T cells following ex vivo PMA/Ionomycin restimulation at study termination across treatment groups, including the combo therapy group with additional FTY720 treatment (combo+F). In combo-treated groups ( E–G ), one data point represents pooled samples from one to six tumors due to small tumor volumes at study termination. Data shown are representative of a single experiment. ( H, I ) Assessment of TCR activation, measured by Nur77-GFP expression, in intratumoral CD4 + and CD8 + T cells, on day 5, 10 and 15 post-treatment across all treatment groups (n=3–4 individual tumors/time point and treatment group. Data shown are representative of a single experiment. ( J ) Representative images of a TDC in a combo-treated tissue. Left: overview with TDC outlined (dotted line, scale bar: 30 µm). Middle: magnified region of the TDC showing CD4 + , CD8 + and Nur77 expression. Right: same region showing CD69 and Nur77 expression (scale bars: 15 µm). Statistical significance in ( E–G ) was determined using one-way ANOVA followed by Dunnett’s post hoc test, with targeted comparisons to the combo group (*p≤0.05, **p≤0.01, ****p≤0.0001). Statistical significance in ( H–I ) was determined using two-way ANOVA with Tukey’s post hoc test (**p≤0.01, ***p≤0.001, ****p≤0.0001). ANOVA, analysis of variance; FAP, fibroblast activation protein; FDR, False Discovery Rate;IFN-γ, interferon-gamma; IL, interleukin; PD-1, programmed cell death protein-1; PMA, phorbol 12-myristate 13-acetate; TCR, T-cell receptor; TDCs, T cell-DC clusters; Th1, T helper cell 1; TNFα, tumor necrosis factor-alpha. Open in a new tab To validate these transcriptional findings at the functional level, we assessed cytokine production by tumor-infiltrating CD4 + and CD8 + T cells through ex vivo stimulation at termination ( Online supplemental figure 9A ). We restimulated dissociated tumor samples with PMA and Ionomycin and measured TNFα and IFN-γ secretion by CD4 + and CD8 + T cells after 4 hours of stimulation. Both CD4 + and CD8 + T cells from combo-treated tumors showed enhanced TNFα and IFN-γ production compared with monotherapy or vehicle control groups, with CD4 + T cells showing higher frequencies of TNFα- and IFN-γ-positive cells compared with CD8 + T cells ( figure 6E,F ). Notably, we observed that the enhanced cytokine response induced by the combo therapy was preserved and even augmented in mice receiving FTY720 ( figure 6G ), demonstrating that the improved functionality of intratumoral T cells is independent of ongoing lymphoid egress, further supporting our previous findings. To further investigate T cell functionality on combo treatment, we evaluated T cell activation directly within the TME using Nur77-GFP reporter mice. In this system, GFP expression is driven by the Nur77 promoter, an early response gene induced on TCR stimulation, thereby providing real-time visualization of TCR signaling within the tumor. 59 Using 3DIP imaging, we analyzed tumor tissues from Nur77-GFP reporter mice bearing KPC tumors 10 days post-treatment initiation. Consistent with our previous observations, combo-treated tumors exhibited significantly higher CD4 + and CD8 + T cell infiltration compared with vehicle controls ( online supplemental figure 9B ). Furthermore, CD4 + T cells in combo-treated tumors showed detectable Nur77-driven GFP expression at day 5 post-treatment, which was significantly increased by days 10 and 15 ( figure 6H ). In contrast, tdLNs showed uniformly high GFP signal across all treatment groups ( online supplemental figure 9C ), indicating that the observed differences in T cell activation between treatment groups are confined to the TME. We further examined CD8 + T cell activation kinetics following combination therapy. While no statistically significant differences were observed between treatment groups at any time point, we noted a trend toward increased Nur77 expression at day 5 post-treatment, which appeared to decline by days 10 and 15 ( figure 6I ). Spatial characterization of the TME further demonstrated that both Nur77 and the early T cell activation marker CD69 60 were expressed within the TDCs that formed following combo treatment on day 10 ( figure 6J ). The simultaneous presence of Nur77-GFP, which can persist in T cells activated during earlier phases of treatment due to GFP protein stability, 59 together with CD69, a marker of recent T cell activation, suggests that TDCs provide a microenvironment that supports both persistent and continuous T cell activation within the tumor. Taken together, the findings provide a detailed cellular and molecular framework explaining the improved antitumor efficacy of combined FAP-CD40 and PD1-IL2v therapy. The combination drives a coordinated immune activation marked by sustained CD4 + T cell signaling, robust Th1 differentiation and formation of intratumoral TDCs, which importantly function independently of B cells or LNs. Ultimately, this work underscores that effective immunotherapy depends not only on the right immune components, but also on their spatial organization within the TME. Discussion In the present study, we demonstrate significant therapeutic benefit by simultaneously targeting DCs and T cell populations in the hard-to-treat KPC tumor model. While FAP-CD40 monotherapy activates intratumoral DCs, it failed to elicit sufficient antitumor activity in this tumor model. Similarly, PD1-IL2v monotherapy, despite specifically delivering IL-2 to antigen-experienced T cells, did not achieve tumor rejection in KPC tumors. However, combining these two distinct and complementary mechanisms—activating DCs with FAP-CD40 while providing activation, proliferation and differentiation signals to T cells—produces synergistic antitumor effects. The enhanced antitumor effect of the combination therapy may be linked to the formation of distinct cDC1-T cell niches (TDCs) within the tumor. These immune structures, found at higher densities in mice showing tumor regression, serve as local hubs that orchestrate antitumor immune responses, potentially contributing to the observed synergistic efficacy. It is essential to differentiate the TDCs therapeutically induced in our study from classical TLS. TLS are characterized by highly organized B and T cell zones and have been linked with favorable treatment outcomes across multiple cancer types. 61 In contrast, TDCs generated by our combo therapy exhibit distinct immune cell composition, marked by the absence of dense B cell infiltrates. Additional TLS-defining features, such as follicular DC networks, were not assessed in this study. Beyond this key distinction, these therapeutically induced TDCs possess further unique features that set them apart from other previously described intratumoral immune aggregates. First, their cellular composition contrasts with that of “stem-immunity hubs” or “cellular triads” reported previously. 26 27 62 While earlier studies highlighted mregDCs interacting with TCF1 + PD-1 + CD8 + T cells or PD-1 + CXCL13 + CD4 + T helper cells, our TDCs were primarily composed of CD4 + and CD8 + T cells colocalizing with Clec9A + CD103 + cDC1s, whereas mregDCs were relatively sparse. The role of mregDCs within TDCs, particularly whether they play any supportive role despite their rather low abundance in these niches, needs further clarification. Unlike the cDC1-CD8 + T cell clusters reported by Meiser et al , 24 which are characterized by MHCII + CCR7 − cDC1s and largely confined to the tumor periphery, TDCs in our study were distributed throughout the tumor, with an enhanced presence within the parenchyma. Their distinct architecture—featuring abundant CD4 + T cells colocalizing with CD8 + T cells—raises important questions about their functional interplay. Depletion studies indicate that both T cell subsets are critical for the therapeutic effect, suggesting a cooperative antitumor mechanism. Furthermore, we observed a reinfiltration of CD4 + and CD8 + T cells into the tumors of depleted groups at the study endpoint. Despite this late reinfiltration, the antitumor response in mice that lacked the targeted T cell subsets at the start of treatment could not be rescued, emphasizing the necessity of maintaining both subsets throughout therapy. Such synergy aligns with the “immune triads” concept proposed by Espinosa et al 25 in which DCs orchestrate the proximity of CD4 + and CD8 + T cells, allowing CD4 + T cells to restore the effector functions of previously exhausted CD8 + T cells. Furthermore, our study expands on the “immune triads” model by suggesting the existence of larger structures that may contain multiple interconnected triads. While not directly tested in our study, we hypothesize that such spatial organization could facilitate CD4 + and CD8 + T cell interactions both within and between triads, thereby creating a highly supportive microenvironment within TDCs. Our scRNA-seq data suggest that the effectiveness of TDCs likely stems from a self-reinforcing feedback loop. Activated cDC1s drive Th1 polarization of CD4 + T cells, as evidenced by enriched IL-12/IL-18 signaling pathways 63 64 and upregulated Th1-associated genes including CD40lg. 65 In turn, CD40L-expressing CD4 + T cells license cDC1s enhancing their IL-12 production and antigen cross-presentation to CD8 + T cells 66 , 68 , which further reinforces Th1 differentiation 69 and CD8 + T cell cytotoxicity. 70 Supporting this model, ex vivo restimulation with PMA/Ionomycin revealed higher frequencies of TNFα + IFN-γ + CD4 + T cells compared with CD8 + T cells in combo-treated tumors, highlighting the prominent contribution of CD4 + T cells to the therapeutic response. Although our study links therapeutically induced TDCs to treatment efficacy, several aspects of their underlying biology remain unresolved. FTY720 experiments show that TDCs sustain antitumor activity independently of ongoing de novo T cell priming in the LNs, raising the question of whether the initial response involves preprimed, antigen-specific T cells or whether the combo therapy can directly engage both naïve and antigen-experienced T cells within the tumor. Additionally, the precise mechanisms driving TDC formation between day 5 and 10 post-treatment, as well as the main factors maintaining these critical niches, require further investigation. Our time-course analyses revealed distinct temporal activation patterns between CD4 + and CD8 + T cells, suggesting a coordinated immune response orchestrated by the combination therapy. Specifically, early CD8 + T cell activation may initiate tumor cell killing but could progress toward functional exhaustion if not sustained. In contrast, prolonged CD4 + T cell activation may provide critical cytokine support and costimulatory signals within these hubs, thereby promoting CD8 + T cell survival and maintaining effective antitumor responses. Future validation of this combination approach and the associated TDC formation across diverse and clinically relevant preclinical tumor models will be crucial. The potential translation of FAP-CD40 and PD1-IL2v combination therapy to humans is particularly promising given the conserved mechanisms of spatial immune organization. Multiple studies have demonstrated that the spatial DC-T cell clustering correlates with improved outcomes across various human cancers. 25 26 71 Furthermore, recent work highlights the role of cDC1s in orchestrating TLS formation via antigen presentation and CD40 signaling, 72 suggesting that FAP-CD40 could further enhance their critical function in tumors. Complementing these findings, Jansen et al identified stem-like CD8 + T cells residing within APC niches in human cancers. 23 Their work demonstrated that insufficient stimulation of the stem-like T cells can lead to declining T cell responses and impaired generation of effector cells. In this context, the PD1-IL2v component of our combination therapy could provide the necessary survival and proliferation signals needed for effective antitumor immunity. Together, these insights strengthen the mechanistic rationale for testing this combination approach in human cancers. Collectively, our study demonstrates that combining FAP-CD40 with PD1-IL2v drives the formation of TDCs within immunologically cold tumors, offering a promising strategy for immunotherapy-resistant solid tumors. By promoting overall immune cell infiltration and specifically inducing TDC formation, this combination not only represents a potential novel treatment by itself but may also sensitize previously refractory tumors to standard treatments by harnessing and amplifying the body’s own antitumor immune responses. Supplementary material online supplemental figure 1 jitc-14-5-s001.pdf (3.1MB, pdf) DOI: 10.1136/jitc-2025-014620 Acknowledgements We thank Wilson Macedo, Jacqueline Hirth, Marie-Lena Mörbitz, Birte Appelt, Ahmet Varol, Marine LeClech and Andrzej Sobieniecki for their support in the animal facility. We thank Caroline Waltzinger for her help with cell sorting and Vera Griesser for her support in sequencing of the tumor samples. We also thank Philipp Fröbel, Cornelia Heuberger, Anne-Claire Zagdoun and Maria Antsiferova for their help with tumor tissue processing. Illustrations in the graphical abstract and in Fig. 2A, 4A, 5A, Supplementary Fig. 2B and Supplementary Fig. 9A were created with BioRender.com. Additionally, we acknowledge the use of AI editing tools for assistance in the preparation and editing of this manuscript. Footnotes Funding: This study was funded by F Hoffmann-La Roche. As employees of the funder, the authors were involved in study design, data collection, analysis, interpretation, and manuscript preparation. The authors confirm that these affiliations did not influence the study results or their interpretation. Provenance and peer review: Not commissioned; externally peer reviewed. Patient consent for publication: Not applicable. Ethics approval: Not applicable. Data availability free text: The single-cell RNA sequencing data generated in this study are available in the BioStudies database under accession number E-MTAB-16500 (see reference73). All other data supporting the findings of this study are available from the corresponding author upon reasonable request. Data availability statement Data are available in a public, open access repository. Data are available upon reasonable request. References 1. June CH, Sadelain M. Chimeric Antigen Receptor Therapy. N Engl J Med. 2018;379:64–73. doi: 10.1056/NEJMra1706169. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 2. Rosenberg SA. IL-2: the first effective immunotherapy for human cancer. J Immunol. 2014;192:5451–8. doi: 10.4049/jimmunol.1490019. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 3. Sahin U, Türeci Ö. Personalized vaccines for cancer immunotherapy. Science. 2018;359:1355–60. doi: 10.1126/science.aar7112. [ DOI ] [ PubMed ] [ Google Scholar ] 4. Sharma P, Allison JP. Immune checkpoint targeting in cancer therapy: toward combination strategies with curative potential. Cell. 2015;161:205–14. doi: 10.1016/j.cell.2015.03.030. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 5. Fares CM, Van Allen EM, Drake CG, et al. Mechanisms of Resistance to Immune Checkpoint Blockade: Why Does Checkpoint Inhibitor Immunotherapy Not Work for All Patients? Am Soc Clin Oncol Educ Book. 2019;39:147–64. doi: 10.1200/EDBK_240837. [ DOI ] [ PubMed ] [ Google Scholar ] 6. Wei SC, Duffy CR, Allison JP. Fundamental Mechanisms of Immune Checkpoint Blockade Therapy. Cancer Discov. 2018;8:1069–86. doi: 10.1158/2159-8290.CD-18-0367. [ DOI ] [ PubMed ] [ Google Scholar ] 7. Brahmer JR, Tykodi SS, Chow LQM, et al. Safety and Activity of Anti–PD-L1 Antibody in Patients with Advanced Cancer. N Engl J Med. 2012;366:2455–65. doi: 10.1056/NEJMoa1200694. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 8. Morrison AH, Byrne KT, Vonderheide RH. Immunotherapy and Prevention of Pancreatic Cancer. Trends Cancer. 2018;4:418–28. doi: 10.1016/j.trecan.2018.04.001. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 9. Ullman NA, Burchard PR, Dunne RF, et al. Immunologic Strategies in Pancreatic Cancer: Making Cold Tumors Hot. J Clin Oncol. 2022;40:2789–805. doi: 10.1200/jco.21.02616. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 10. Ene-Obong A, Clear AJ, Watt J, et al. Activated pancreatic stellate cells sequester CD8+ T cells to reduce their infiltration of the juxtatumoral compartment of pancreatic ductal adenocarcinoma. Gastroenterology. 2013;145:1121–32. doi: 10.1053/j.gastro.2013.07.025. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 11. Goulart MR, Stasinos K, Fincham REA, et al. T cells in pancreatic cancer stroma. World J Gastroenterol. 2021;27:7956–68. doi: 10.3748/wjg.v27.i46.7956. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 12. Lin JH, Huffman AP, Wattenberg MM, et al. Type 1 conventional dendritic cells are systemically dysregulated early in pancreatic carcinogenesis. J Exp Med. 2020;217:e20190673. doi: 10.1084/jem.20190673. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 13. Spranger S, Dai D, Horton B, et al. Tumor-Residing Batf3 Dendritic Cells Are Required for Effector T Cell Trafficking and Adoptive T Cell Therapy. Cancer Cell. 2017;31:711–23. doi: 10.1016/j.ccell.2017.04.003. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 14. Hegde S, Krisnawan VE, Herzog BH, et al. Dendritic Cell Paucity Leads to Dysfunctional Immune Surveillance in Pancreatic Cancer. Cancer Cell. 2020;37:289–307. doi: 10.1016/j.ccell.2020.02.008. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 15. Hingorani SR, Wang L, Multani AS, et al. Trp53R172H and KrasG12D cooperate to promote chromosomal instability and widely metastatic pancreatic ductal adenocarcinoma in mice. Cancer Cell. 2005;7:469–83. doi: 10.1016/j.ccr.2005.04.023. [ DOI ] [ PubMed ] [ Google Scholar ] 16. Vonderheide RH, Bear AS. Tumor-Derived Myeloid Cell Chemoattractants and T Cell Exclusion in Pancreatic Cancer. Front Immunol. 2020;11:605619. doi: 10.3389/fimmu.2020.605619. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 17. Li J, Byrne KT, Yan F, et al. Tumor Cell-Intrinsic Factors Underlie Heterogeneity of Immune Cell Infiltration and Response to Immunotherapy. Immunity. 2018;49:178–93. doi: 10.1016/j.immuni.2018.06.006. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 18. Hogg GD, Weinstein AG, Kingston NL, et al. Combined Flt3L and CD40 agonism restores dendritic cell-driven T cell immunity in pancreatic cancer. Sci Immunol. 2025;10:eadp3978. doi: 10.1126/sciimmunol.adp3978. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 19. Sum E, Rapp M, Fröbel P, et al. Fibroblast Activation Protein α-Targeted CD40 Agonism Abrogates Systemic Toxicity and Enables Administration of High Doses to Induce Effective Antitumor Immunity. Clin Cancer Res. 2021;27:4036–53. doi: 10.1158/1078-0432.CCR-20-4001. [ DOI ] [ PubMed ] [ Google Scholar ] 20. Steele NG, Carpenter ES, Kemp SB, et al. Multimodal Mapping of the Tumor and Peripheral Blood Immune Landscape in Human Pancreatic Cancer. Nat Cancer . 2020;1:1097–112. doi: 10.1038/s43018-020-00121-4. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Codarri Deak L, Nicolini V, Hashimoto M, et al. PD-1-cis IL-2R agonism yields better effectors from stem-like CD8+ T cells. Nature New Biol. 2022;610:161–72. doi: 10.1038/s41586-022-05192-0. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Pittet MJ, Di Pilato M, Garris C, et al. Dendritic cells as shepherds of T cell immunity in cancer. Immunity. 2023;56:2218–30. doi: 10.1016/j.immuni.2023.08.014. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 23. Jansen CS, Prokhnevska N, Master VA, et al. An intra-tumoral niche maintains and differentiates stem-like CD8 T cells. Nature New Biol. 2019;576:465–70. doi: 10.1038/s41586-019-1836-5. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. Meiser P, Knolle MA, Hirschberger A, et al. A distinct stimulatory cDC1 subpopulation amplifies CD8+ T cell responses in tumors for protective anti-cancer immunity. Cancer Cell. 2023;41:1498–515. doi: 10.1016/j.ccell.2023.06.008. [ DOI ] [ PubMed ] [ Google Scholar ] 25. Espinosa-Carrasco G, Chiu E, Scrivo A, et al. Intratumoral immune triads are required for immunotherapy-mediated elimination of solid tumors. Cancer Cell. 2024;42:1202–16. doi: 10.1016/j.ccell.2024.05.025. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 26. Chen JH, Nieman LT, Spurrell M, et al. Human lung cancer harbors spatially organized stem-immunity hubs associated with response to immunotherapy. Nat Immunol. 2024;25:644–58. doi: 10.1038/s41590-024-01792-2. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 27. Magen A, Hamon P, Fiaschi N, et al. Intratumoral dendritic cell–CD4+ T helper cell niches enable CD8+ T cell differentiation following PD-1 blockade in hepatocellular carcinoma. Nat Med. 2023;29:1389–99. doi: 10.1038/s41591-023-02345-0. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 28. Stoltzfus CR, Sivakumar R, Kunz L, et al. Multi-Parameter Quantitative Imaging of Tumor Microenvironments Reveals Perivascular Immune Niches Associated With Anti-Tumor Immunity. Front Immunol. 2021;12:726492. doi: 10.3389/fimmu.2021.726492. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 29. Kersemans V, Cornelissen B, Allen PD, et al. Subcutaneous tumor volume measurement in the awake, manually restrained mouse using MRI. J Magn Reson Imaging. 2013;37:1499–504. doi: 10.1002/jmri.23829. [ DOI ] [ PubMed ] [ Google Scholar ] 30. Tomayko MM, Reynolds CP. Determination of subcutaneous tumor size in athymic (nude) mice. Cancer Chemother Pharmacol. 1989;24:148–54. doi: 10.1007/BF00300234. [ DOI ] [ PubMed ] [ Google Scholar ] 31. Stoltzfus CR, Filipek J, Gern BH, et al. CytoMAP: A Spatial Analysis Toolbox Reveals Features of Myeloid Cell Organization in Lymphoid Tissues. Cell Rep. 2020;31:107523. doi: 10.1016/j.celrep.2020.107523. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 32. Fischer DS, Schaar AC, Theis FJ. Modeling intercellular communication in tissues using spatial graphs of cells. Nat Biotechnol. 2023;41:332–6. doi: 10.1038/s41587-022-01467-z. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 33. Mädler SC, Julien-Laferriere A, Wyss L, et al. Besca, a single-cell transcriptomics analysis toolkit to accelerate translational research. NAR Genom Bioinform. 2021;3:lqab102. doi: 10.1093/nargab/lqab102. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 34. Becht E, McInnes L, Healy J, et al. Dimensionality reduction for visualizing single-cell data using UMAP. Nat Biotechnol. 2019;37:38–44. doi: 10.1038/nbt.4314. [ DOI ] [ PubMed ] [ Google Scholar ] 35. Traag VA, Waltman L, van EN. From Louvain to Leiden: guaranteeing well-connected communities. Sci Rep. 2019;9:5233. doi: 10.1038/s41598-019-41695-z. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 36. Germain P-L, Lun A, Garcia Meixide C, et al. Doublet identification in single-cell sequencing data using scDblFinder. F1000Res. 2021;10:979. doi: 10.12688/f1000research.73600.2. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 37. Robinson MD, Oshlack A. A scaling normalization method for differential expression analysis of RNA-seq data. Genome Biol. 2010;11:R25. doi: 10.1186/gb-2010-11-3-r25. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 38. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010;26:139–40. doi: 10.1093/bioinformatics/btp616. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 39. Risso D, Ngai J, Speed TP, et al. Normalization of RNA-seq data using factor analysis of control genes or samples. Nat Biotechnol. 2014;32:896–902. doi: 10.1038/nbt.2931. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 40. Chu Y, Dai E, Li Y, et al. Pan-cancer T cell atlas links a cellular stress response state to immunotherapy resistance. Nat Med. 2023;29:1550–62. doi: 10.1038/s41591-023-02371-y. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 41. Luckheeram RV, Zhou R, Verma AD, et al. CD4+T Cells: Differentiation and Functions. J Immunol Res. 2012:925135. doi: 10.1155/2012/925135. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 42. Ciernik L, Kraft A, Barkmann F, et al. ANS: Adjusted Neighborhood Scoring to improve gene signature-based cell annotation in single-cell RNA-seq data. bioRxiv. 2025 doi: 10.1101/2023.09.20.558114. [ DOI ] [ Google Scholar ] 43. Verdijk P, van Veelen PA, de Ru AH, et al. Morphological changes during dendritic cell maturation correlate with cofilin activation and translocation to the cell membrane. Eur J Immunol. 2004;34:156–64. doi: 10.1002/eji.200324241. [ DOI ] [ PubMed ] [ Google Scholar ] 44. Cruz-Matías I, Ayala D, Hiller D, et al. Sphericity and roundness computation for particles using the extreme vertices model. J Comput Sci. 2019;30:28–40. doi: 10.1016/j.jocs.2018.11.005. [ DOI ] [ Google Scholar ] 45. Cardenas MA, Prokhnevska N, Sobierajska E, et al. Differentiation fate of a stem-like CD4 T cell controls immunity to cancer. Nature New Biol. 2024;636:224–32. doi: 10.1038/s41586-024-08076-7. [ DOI ] [ PubMed ] [ Google Scholar ] 46. Im SJ, Hashimoto M, Gerner MY, et al. Defining CD8+ T cells that provide the proliferative burst after PD-1 therapy. Nature New Biol. 2016;537:417–21. doi: 10.1038/nature19330. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 47. Siddiqui I, Schaeuble K, Chennupati V, et al. Intratumoral Tcf1+PD-1+CD8+ T Cells with Stem-like Properties Promote Tumor Control in Response to Vaccination and Checkpoint Blockade Immunotherapy. Immunity. 2019;50:195–211. doi: 10.1016/j.immuni.2018.12.021. [ DOI ] [ PubMed ] [ Google Scholar ] 48. Brinkmann V, Cyster JG, Hla T. FTY720: sphingosine 1-phosphate receptor-1 in the control of lymphocyte egress and endothelial barrier function. Am J Transplant. 2004;4:1019–25. doi: 10.1111/j.1600-6143.2004.00476.x. [ DOI ] [ PubMed ] [ Google Scholar ] 49. Szabo SJ, Kim ST, Costa GL, et al. A novel transcription factor, T-bet, directs Th1 lineage commitment. Cell. 2000;100:655–69. doi: 10.1016/s0092-8674(00)80702-3. [ DOI ] [ PubMed ] [ Google Scholar ] 50. Thieu VT, Yu Q, Chang H-C, et al. Signal transducer and activator of transcription 4 is required for the transcription factor T-bet to promote T helper 1 cell-fate determination. Immunity. 2008;29:679–90. doi: 10.1016/j.immuni.2008.08.017. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 51. Nakahira M, Ahn H-J, Park W-R, et al. Synergy of IL-12 and IL-18 for IFN-γ Gene Expression: IL-12-Induced STAT4 Contributes to IFN-γ Promoter Activation by Up-Regulating the Binding Activity of IL-18-Induced Activator Protein 1. J Immunol. 2002;168:1146–53. doi: 10.4049/jimmunol.168.3.1146. [ DOI ] [ PubMed ] [ Google Scholar ] 52. Linterman MA, Denton AE, Divekar DP, et al. CD28 expression is required after T cell priming for helper T cell responses and protective immunity to infection. Elife. 2014;3:e03180. doi: 10.7554/eLife.03180. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 53. Liao W, Lin JX, Leonard WJ. IL-2 family cytokines: new insights into the complex roles of IL-2 as a broad regulator of T helper cell differentiation. Curr Opin Immunol. 2011;23:598–604. doi: 10.1016/j.coi.2011.08.003. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 54. Kim CH, Kunkel EJ, Boisvert J, et al. Bonzo/CXCR6 expression defines type 1–polarized T-cell subsets with extralymphoid tissue homing potential. J Clin Invest. 2001;107:595–601. doi: 10.1172/JCI11902. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 55. Loetscher P, Uguccioni M, Bordoli L, et al. CCR5 is characteristic of Th1 lymphocytes. Nature New Biol. 1998;391:344–5. doi: 10.1038/34814. [ DOI ] [ PubMed ] [ Google Scholar ] 56. Kumar A, Humphreys TD, Kremer KN, et al. CXCR4 Physically Associates with the T Cell Receptor to Signal in T Cells. Immunity. 2006;25:213–24. doi: 10.1016/j.immuni.2006.06.015. [ DOI ] [ PubMed ] [ Google Scholar ] 57. Overstreet MG, Gaylo A, Angermann BR, et al. Inflammation-induced interstitial migration of effector CD4+ T cells is dependent on integrin αV. Nat Immunol. 2013;14:949–58. doi: 10.1038/ni.2682. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 58. Wang Z, Shang J, Qiu Y, et al. Suppression of the METTL3-m6A-integrin β1 axis by extracellular acidification impairs T cell infiltration and antitumor activity. Cell Rep. 2024;43:113796. doi: 10.1016/j.celrep.2024.113796. [ DOI ] [ PubMed ] [ Google Scholar ] 59. Moran AE, Holzapfel KL, Xing Y, et al. T cell receptor signal strength in Treg and iNKT cell development demonstrated by a novel fluorescent reporter mouse. J Exp Med. 2011;208:1279–89. doi: 10.1084/jem.20110308. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 60. Sancho D, Gómez M, Sánchez-Madrid F. CD69 is an immunoregulatory molecule induced following activation. Trends Immunol. 2005;26:136–40. doi: 10.1016/j.it.2004.12.006. [ DOI ] [ PubMed ] [ Google Scholar ] 61. Vanhersecke L, Brunet M, Guégan J-P, et al. Mature tertiary lymphoid structures predict immune checkpoint inhibitor efficacy in solid tumors independently of PD-L1 expression. Nat Cancer . 2021;2:794–802. doi: 10.1038/s43018-021-00232-6. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 62. Cohen M, Giladi A, Barboy O, et al. The interaction of CD4+ helper T cells with dendritic cells shapes the tumor microenvironment and immune checkpoint blockade response. Nat Cancer . 2022;3:303–17. doi: 10.1038/s43018-022-00338-5. [ DOI ] [ PubMed ] [ Google Scholar ] 63. Macatonia SE, Hosken NA, Litton M, et al. Dendritic cells produce IL-12 and direct the development of Th1 cells from naive CD4+ T cells. J Immunol. 1995;154:5071–9. doi: 10.4049/jimmunol.154.10.5071. [ DOI ] [ PubMed ] [ Google Scholar ] 64. Nakanishi K, Yoshimoto T, Tsutsui H, et al. Interleukin-18 is a unique cytokine that stimulates both Th1 and Th2 responses depending on its cytokine milieu. Cytokine Growth Factor Rev. 2001;12:53–72. doi: 10.1016/s1359-6101(00)00015-0. [ DOI ] [ PubMed ] [ Google Scholar ] 65. Cella M, Scheidegger D, Palmer-Lehmann K, et al. Ligation of CD40 on dendritic cells triggers production of high levels of interleukin-12 and enhances T cell stimulatory capacity: T-T help via APC activation. J Exp Med. 1996;184:747–52. doi: 10.1084/jem.184.2.747. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 66. Ferris ST, Durai V, Wu R, et al. cDC1 prime and are licensed by CD4+ T cells to induce anti-tumour immunity. Nature New Biol. 2020;584:624–9. doi: 10.1038/s41586-020-2611-3. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 67. Borst J, Ahrends T, Bąbała N, et al. CD4+ T cell help in cancer immunology and immunotherapy. Nat Rev Immunol. 2018;18:635–47. doi: 10.1038/s41577-018-0044-0. [ DOI ] [ PubMed ] [ Google Scholar ] 68. Hildner K, Edelson BT, Purtha WE, et al. Batf3 Deficiency Reveals a Critical Role for CD8α+ Dendritic Cells in Cytotoxic T Cell Immunity. Science. 2008;322:1097–100. doi: 10.1126/science.1164206. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 69. Koch F, Stanzl U, Jennewein P, et al. High level IL-12 production by murine dendritic cells: upregulation via MHC class II and CD40 molecules and downregulation by IL-4 and IL-10. J Exp Med. 1996;184:741–6. doi: 10.1084/jem.184.2.741. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 70. Ruffell B, Chang-Strachan D, Chan V, et al. Macrophage IL-10 blocks CD8+ T cell-dependent responses to chemotherapy by suppressing IL-12 expression in intratumoral dendritic cells. Cancer Cell. 2014;26:623–37. doi: 10.1016/j.ccell.2014.09.006. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 71. Lee CYC, Kennedy BC, Richoz N, et al. Tumour-retained activated CCR7+ dendritic cells are heterogeneous and regulate local anti-tumour cytolytic activity. Nat Commun. 2024;15:682. doi: 10.1038/s41467-024-44787-1. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 72. Mattiuz R, Boumelha J, Hamon P, et al. Dendritic cells type 1 control the formation, maintenance, and function of tertiary lymphoid structures in cancer. bioRxiv. 2024 doi: 10.1101/2024.12.27.628014. [ DOI ] [ Google Scholar ] 73. Nguyen TT, et al. Data from: FAP-CD40 and PD1-IL2v combination therapy reprograms immunologically cold tumors through de novo intratumoral T cell-dendritic cell clusters. BioStudies Database. 2026 doi: 10.1136/jitc-2025-014620. https://www.ebi.ac.uk/biostudies/arrayexpress/studies/E-MTAB-16500 Available. [ 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 online supplemental figure 1 jitc-14-5-s001.pdf (3.1MB, pdf) DOI: 10.1136/jitc-2025-014620 Data Availability Statement Data are available in a public, open access repository. Data are available upon reasonable request. Articles from Journal for Immunotherapy of Cancer are provided here courtesy of BMJ Publishing Group ACTIONS View on publisher site PDF (15.5 MB) Cite Collections Permalink RESOURCES Similar articles Cited by other articles Links to NCBI Databases Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• The tumor-sentinel lymph node immuno-migratome reveals CCR7⁺ dendritic cells drive response to sequenced immunoradiotherapy#996252
• Induced-T Cell Like NK Cellular Immunotherapy for Cancer Lack of MHC-I#633726
• Controlled Cold Exposure Combined With PD-1/PD-L1 Immunotherapy in Solid Tumors (NIVALIS)#272361
• CD8(+) T cell differentiation into NK-like effector cells drives transplant rejection.#295207
• Tumor Gene Expression Heterogeneity Underlies T Cell Evasion#637498