Engineered artificial antigen-presenting cells for tumor immunotherapy: Design strategies, clinical applications, and translational challenges Authors: Haojie Cai, Yaqian Zhao, Yifei Cai, Xiaomei Luo, Jianxiao Li, Yongsong Wu, Jiani Yang, Chao Wang, Yue Zhang, Hui Ding, Shanshan Cheng, Yu Wang Published: 2026 ABSTRACT AbstractImmunotherapy has long played a pivotal role in cancer treatment, and antigen-presenting cell (APC)-based immunotherapy represents a promising strategy, though its full potential remains to be realized. This therapy stimulates T cell activation and proliferation through APCs, producing therapeutic effects. However, the use of natural APCs in immunotherapy faces several limitations, including inefficient T cell expansion, suboptimal targeting, and high treatment costs. Under such circumstances, artificial antigen-presenting cells (aAPCs) have garnered increasing interest. This review provides an overview of recent advancements in aAPCs research, systematically classifies aAPCs and critically evaluates their advantages and limitations. The underlying mechanisms of aAPCs, including activation and proliferation of T cells, strategies to overcome the immunosuppressive tumor microenvironment (TME), targeted antigen delivery and combination therapy, are discussed in detail. Their clinical applications are also debated. Existing challenges, including the absence of standardized protocols and regulatory frameworks, are highlighted. Furthermore, future directions in aAPCs technology are explored, including but not limited to the development of multifunctional aAPCs and the integration of artificial intelligence (AI) to assist in the design and optimization. Finally, to overcome these limitations, this review advocates for strengthened multidisciplinary collaboration and the adoption of emerging technologies.Keywords: Artificial antigen-presenting cells, Immunotherapy, T cells, Cancer Introduction Cancer remains a major global health concern [1] due to its profound impact on human life [2]. With the continuous advancement of modern medicine, numerous studies have been conducted to explore effective therapeutic strategies. A wide range of treatments has been proposed by researchers and clinicians worldwide, including surgical resection [3], chemotherapy [4], radiotherapy [5], and gene therapy [6]. Among these, immunotherapy, which harnesses the body's own immune system, has emerged as a central modality owing to its high specificity and broad applicability. Particularly in tumor immunotherapy, T cell-mediated immune responses play a central role, and the initiation of T cell activation strongly depends on antigen-presenting cells (APCs). However, natural APCs still face critical drawbacks in clinical applications, highlighting the need for alternative strategies. Multiple forms of immunotherapy have been developed, with chimeric antigen receptor T cell (CAR-T) therapy receiving particular attention [7]. In this approach, T cells are isolated from a patient's blood or tumor tissue and genetically modified to express a chimeric antigen receptor (CAR) that targets tumor-specific antigens, such as CD19. These engineered T cells are subsequently expanded ex vivo and reinfused into the patient, enabling targeted elimination of cancer cells. Substantial progress has recently been achieved in this area [8]. Besides, immune checkpoint blockade (ICB) represents another widely adopted immunotherapeutic strategy. For example, inhibitors of programmed cell death protein 1 (PD-1) and programmed death-ligand 1 (PD-L1) function by blocking the interaction between PD-L1 on tumor cells and PD-1 on T cells [9]. This blockade prevents T cell inactivation and reactivates the patient's immune system, facilitating sustained anti-tumor activity. Notably, both CAR-T therapy and ICB ultimately rely on efficient T cell activation, which is initiated and regulated by APCs. In addition to CAR-T and ICB, other important immunotherapeutic strategies include adoptive cell therapy (ACT), tumor-infiltrating lymphocyte (TIL) therapy, cancer vaccines, NK cell therapy, and cytokine therapy. All of these approaches rely on efficient antigen presentation and T cell activation, highlighting the key value of aAPCs in broad tumor immunotherapy scenarios. Given the central role of APCs in T cell activation and the limitations of natural APCs, the development of artificial antigen-presenting cells (aAPCs) has emerged as a promising strategy to enhance the efficacy of diverse immunotherapeutic approaches. Recently, therapies based on aAPCs have attracted growing attention as promising supplements to the current landscape of immunotherapeutic approaches. Antigen-presenting cells (APCs) and tumor immunotherapy APCs are a specialized group of immune cells that capture, process, and present antigens to T cells, thereby initiating and regulating adaptive immune responses. In contrast to natural APCs, aAPCs are engineered platforms—ranging from cell-based constructs to synthetic biomaterials—designed to mimic the key features of natural APCs, offering scalable, customizable, and often more efficient alternatives for T cell stimulation in cancer immunotherapy. APCs—including dendritic cells (DCs), macrophages, and B cells—play a pivotal role in capturing and presenting tumor antigens. These antigens may derive from tumor-specific surface proteins or intracellular components released by malignant cells. Antigen uptake is mediated via phagocytosis, macropinocytosis, or receptor-dependent endocytosis. Once internalized, the antigens are processed and loaded onto major histocompatibility complex (MHC) molecules to form peptide-MHC (pMHC) complexes. MHC class I molecules predominantly present endogenous peptides to CD8+ T cells, while MHC class II molecules present exogenous peptides to CD4+ T cells. Engagement of pMHC with the T cell receptor (TCR) initiates T cell activation. Beyond antigen presentation, APCs express co-stimulatory molecules, such as CD80 (B7.1) and CD86 (B7.2), which interact with CD28 on T cells, delivering the essential second signal for full T cell activation. In the absence of this signal, T cells may become anergic or undergo apoptosis. APCs also secrete cytokines—including interleukin-12 (IL-12) and interleukin-18 (IL-18)—that support T cell proliferation and differentiation. Consequently, activated T cells undergo clonal expansion. CD8+T cells differentiate into cytotoxic T lymphocytes (CTLs) capable of directly eliminating tumor cells, while CD4+ T cells become helper T cells (Th cells) that orchestrate immune responses through cytokine secretion. Following activation, T cells migrate toward tumor sites in response to chemokines such as CCL2 and CXCL10. Adhesion molecules, including lymphocyte function-associated antigen-1 (LFA-1), facilitate transmigration across the endothelium. Upon reaching the tumor microenvironment (TME), CTLs execute cytotoxicity via multiple mechanisms: the release of perforin and granzymes; Fas ligand (FasL)-mediated apoptosis through interaction with Fas on tumor cells; or secretion of interferon-gamma (IFN-γ) and interferon-alpha (IFN-α) to induce immunogenic cell death. Additionally, cytokines such as IL-2 and IL-12 amplify anti-tumor immunity by activating macrophages and natural killer (NK) cells. Overall, these coordinated processes enable APCs to initiate and sustain T cell-mediated immune responses, forming a cornerstone of anti-tumor immunotherapy (Fig. 1) [10]. Natural T cell activation and mechanism of cytotoxic T lymphocyte in tumor immunotherapy (By Figdraw). Based on 3 signals, naive T cells (CD8+ T cells are shown here as an example, while the mechanism for CD4+ T cells is similar) are completely activated. First, peptide-major histocompatibility complex (pMHC) on antigen-presenting cells (APCs) interacts with T cell receptors (TCRs) on T cells. Second, costimulatory ligands (e.g., CD80 or CD86) on APCs will interact with costimulatory receptors on T cells (e.g., CD28). In addition, some cytokines, including interleukin-12 (IL-12) and IL-16, are secreted and exert their effects. T cells then proliferate and differentiate into different types, including cytotoxic T lymphocytes (CTL). Attracted by chemokines (e.g., CCL2, CXCL10), CTLs are targeted and transferred into the tumor microenvironment (TME) and kill tumor cells via pathways such as Fas/FasL, ultimately leading to tumor cell apoptosis. This figure is based on the mechanisms described in Chopp et al.’s work [10]. Limitation of natural APCs However, conventional approaches utilizing natural APCs for anti-tumor immunotherapy exhibit several intrinsic limitations. First, one strategy employs tumor vaccines, including antigen and other adjuvants, to stimulate APC-mediated T cell activation. APCs take up antigenic components from tumor vaccines, process them into pMHC complexes, and display them on their surface. They then collaborate with the adjuvant components in the vaccine to promote T cell activation. However, the interaction between vaccines and APCs is largely inefficient, often resulting in suboptimal activation, proliferation, and differentiation of CD8+ T cells [11]. Significant research efforts have focused on addressing this question [[12], [13], [14]]. Alternatively, natural APCs can be used ex vivo to activate and expand T cells. While conceptually effective, this approach is labor-intensive and yields limited cell production [15]. Moreover, CTLs generated in vitro show significantly reduced survival post-infusion [16], leading to insufficient CTLs infiltration into the TME and diminished antitumor responses. To overcome these limitations, increasing attention has turned toward the development of advanced biomaterials. In this context, aAPCs have emerged as a promising platform for enhancing T cell-mediated cancer immunotherapy. Advantages and applications of aAPCs A diverse array of aAPCs has been developed to address the inherent limitations of natural APCs-based strategies. These engineered platforms exhibit numerous advantages that enhance T cell activation, proliferation, and antitumor efficacy. For example, Oelke et al. established a bead-based aAPC system coated with HLA-Ig and anti-CD28 antibody, which achieved efficient ex vivo induction and expansion of CMV- and melanoma-specific CD8+ CTLs and enabled the generated CTLs to recognize endogenously processed tumor antigens [17]. In 2017, Zhang et al. developed a multifunctional aAPC incorporating magnetic nanoclusters (MNCs), leukocyte-derived membranes, co-stimulatory molecules, and pMHC complexes. This construct effectively overcame key shortcomings of conventional methods–notably low T cell expansion efficiency and limited antigen specificity [18]. That same year, Kosmides et al. introduced a biodegradable aAPC co-delivered with anti-PD-1 antibodies, demonstrating a synergistic approach that integrates combination immunotherapy, biodegradability, and enhanced T cell functionality [19]. In a clinical context, Butler et al. conducted a trial using aAPCs in patients with advanced melanoma, with results indicating potential therapeutic benefit [20]. In summary, these studies show the versatility, modularity, and translational promise of aAPCs. To summarize, given the limitations of traditional natural APCs-based therapies and based on the booming development of tumor immunotherapy, there is a growing interest in finding artificial alternatives to natural APCs. The aAPCs are increasingly recognized as pivotal platforms in this evolving landscape. This review will summarize the latest developments in aAPCs design, classify current types, explore their underlying mechanisms, highlight clinical applications, and discuss existing challenges and future directions in the field. Classifications of aAPCs Based on their different sources and synthesis strategies, aAPCs can be primarily categorized into three types: cell-based aAPCs, particle-based aAPCs and aAPC platform systems. The following sections will provide a detailed introduction to each category (Fig. 2). To facilitate a systematic understanding and guide platform selection, we summarize the key characteristics, advantages, and limitations of each aAPC category in Table 1. Classifications of artificial antigen-presenting cells (aAPCs). aAPCs can be categorized into three types: cell-based aAPCs, particle-based aAPCs, and aAPC platform systems. Cell-based aAPCs are primarily constructed via (1) modified cell lines with antigen-presenting functions and (2) enhanced natural APCs. Particle-based aAPCs can be further subdivided into natural particle-based aAPCs, synthetic particle-based aAPCs, and hybrid particle-based aAPCs. Natural particle-based aAPCs primarily refer to vesicles constructed from DC cells or other cells. Synthetic particle-based aAPCs are categorized into liposomes, PLGA, latex, emerging micro-/nanoparticle materials, magnetic beads, and DNA-based aAPCs. Hybrid particle-based aAPCs are primarily constructed through two approaches: (1) DC cell membrane-coated nanoparticles and (2) engineered cell membranes incorporating nanoparticles. Regarding aAPCs platform systems, the main construction concepts are biodegradable scaffolds and hydrogel-based aAPCs platform systems. The classification framework is based on the categorization discussed in Section 2. Note: The tubular structure is a schematic representation of a generic biodegradable scaffold and does not refer to any specific material used in this study. Table 1Comparison of Key aAPC Platforms for Tumor Immunotherapy.Table 1Platform CategoryCell-BasedNatural Particle-BasedSynthetic Particle-BasedHybrid Particle-BasedPlatform SystemsRepresentative SystemsK562, NIH-3T3DC-derived exosomesPLGA, liposomes, magnetic beadsDC membrane-coated nanoparticles3D hydrogels, macroporous scaffoldsEx Vivo Suitability+++++++++++++++++++++In Vivo Suitability+++++++++++++++++++Activation EfficiencyHighModerate to HighHigh (tunable)HighHighBiosafetyModerateHighHigh (degradable)/Moderate (non-degradable)HighHighStabilityHighModerateHighModerate to HighHighKey AdvantagesWell-established protocols; excellent scalability; phenotypic stabilityHigh biocompatibility; low immunogenicity; prolonged circulationPrecise control over ligand density; scalable manufacturingSynergistic targeting and drug delivery; immune evasionMimics lymph node microenvironment; sustained local delivery; mechanical signalingKey LimitationsForeign gene integration risk; requires irradiation for in vivo useLow yield; complex purification; batch-to-batch variabilitySome materials non-degradable; potential off-target effectsComplex fabrication; limited membrane sourcesInvasive implantation; manufacturing complexityRating scale: +++++ = most favorable; + = least favorable within this comparison. Ratings are relative and based on evidence compiled in the main text. Comparison of Key aAPC Platforms for Tumor Immunotherapy. Rating scale: +++++ = most favorable; + = least favorable within this comparison. Ratings are relative and based on evidence compiled in the main text. To avoid ambiguity in the classification, we adopt a criterion based on structural composition and construction methodology. Cell-based aAPCs are defined by the use of living or modified cells as the core scaffold. Particle-based aAPCs are further divided according to the nature of the particle matrix. Natural particle-based aAPCs consist of biologically derived vesicles (e.g., exosomes) without synthetic cores. Synthetic particle-based aAPCs are fabricated entirely from artificial materials. Hybrid particle-based aAPCs combine a synthetic core (e.g., PLGA nanoparticle) with a functional biological membrane coating, thereby merging the advantages of both natural and synthetic components. Platform systems are distinguished by their macroscale three-dimensional architecture, which provides a structural scaffold for T cell infiltration and sustained signaling, in contrast to the discrete, nanoscale or microscale particles described above. This framework prioritizes structural origin and fabrication strategy as the primary distinguishing factors, ensuring clear differentiation between categories. Cell-based aAPCs Cell-based aAPCs utilize living cells as scaffolds to present T cell activation signals. This category offers well-established genetic engineering protocols, excellent scalability, and stable phenotypic characteristics, making them particularly suitable for ex vivo T cell expansion. However, their in vivo application is limited by potential immunogenicity and the requirement for irradiation prior to use. Cell-based aAPCs are designed using two main strategies: one involves modifying existing cell lines to endow them with antigen-presenting functions, while the other strengthens natural APCs. Modified cell lines with antigen-presenting functions This strategy involves genetically modifying readily available cell lines—most notably K562 and NIH-3T3—to express antigen-presenting molecules (e.g., MHC or HLA) and co-stimulatory ligands (e.g., CD80, 4-1BBL). The engineered cells serve as scalable and reproducible platforms for T cell activation. This design approach is quite common and has a substantial research foundation (Table 2). One of the most well-known examples is the K562 human erythroleukemic cell line. K562 cells are favored for their low immunogenicity, natural expression of adhesion molecules, ease of transfection with molecules like CD32 and 4-1BBL, sensitivity to irradiation, scalability, and phenotypic stability [46]. In 2002, Maus and colleagues reported a study in Nature Biotechnology and they demonstrated that they transfected K562 cells with CD32 and 4-1BBL. This transfection successfully created the K32/4-1BBL cell line, which facilitated sustained expansion of CD8+ T cells [21]. Subsequently, based on the K562 cell line, researchers have developed a wide range of aAPCs, each yielding unique results in terms of T cell activation and expansion [[22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36]]. Table 2Summary of aAPCs based on modified cell lines with antigen-presenting functions for tumor immunotherapy.Table 2Cell LinesPurposeResultsRef.K562 Cell LineTo address low proliferation and long-term culture challenges in CTL expansion, especially for CD8+ T cells, in vitro.To validate 4-1BBL's role in their long-term proliferation, survival, and antigen-specific maintenance.4-1BBL synergizes with CD28 signaling to promote long-term proliferation and reduce apoptosis of CD8+ T cells.Maintains antigen-specific cytotoxicity and TCR diversity, with potential for clinical application.[21]To stably express co-stimulatory molecules/Fc receptors, addressing natural APCs's complex/costly preparation.To verify that aAPC induces CD8+ T cell proliferation/maintains function via co-stimulatory molecules, supporting genetically modified T cell expansion in vitro.aAPC stably expresses multiple co-stimulatory molecules, induces efficient CD8+ T cell proliferation without exogenous cytokines, and better activates CD28−T cells than traditional magnetic beads. Activated T cells maintain phenotype and secrete IL-2/IFN-γ, providing a novel platform for tumor immunotherapy and vaccine development.[22]To improve CAR-T cell production efficiency and overcome monocyte contamination with the development of aAPCs with low-density lipoprotein receptor knockout (aAPC-ΔLDLR).aAPC-ΔLDLR effectively promotes CAR-T expansion, enhances anti-tumor effects, is suitable for multiple cell sources, and simplifies the manufacturing process.[23]To develop a clinical grade aAPC system to efficiently generate long-term viable anti-tumor CTLs in vitro.aAPCs are employed to standardize the preparation of powerful antigen-specific CTLs to support the clinical translation of adoptive immunotherapy.[24]To develop clinical-grade aAPC-amplified clonal γδ T cells and to validate their in vitro/in vivo antitumor activity.aAPC + IL-2/IL-21 effectively expanded polyclonal γδ T cells with broad-spectrum killing ability, and the in vivo efficacy of polyclonal group was better than that of single subpopulation, which supported the clinical application of cancer immunotherapy.[25]To expand γδT cell subpopulations unbiasedly and to analyze the killing mechanism of each subpopulation against neuroblastoma.aAPC achieves unbiased expansion of γδT subsets (Vδ2+, Vδ1+, Vδ1−Vδ2-), Vδ2+ kills GD2+ neuroblastoma via ADCC (dependent on anti-GD2 mAb ch14.18), while Vδ1+ and double-negative subsets kill the same target via potent AIC, offering a new strategy for its immunotherapy.[26]To develop aAPC expressing HLA-DP4 and to validate its ability to induce HLA-DP4-restricted anti-survivin CD4+ T cell responses.To investigate the phenotype, function, in vitro lifespan, and tumor cell recognition of anti-survivin CD4+ T cells to enhance the efficacy of anti-survivin cancer immunotherapy.HLA-DP4-aAPC induces long-surviving anti-survivin CD4+ T cells with both Th1 and Th2 phenotypes, possessing tumor-recognizing ability and long-lasting efficacy in vitro.[27]To construct a multifunctional aAPC (EpCAM/chemokine/co-stimulatory molecule).To investigate the therapeutic effect of aAPC combined with CAR-T and cytokines on solid tumors.aAPC promotes CAR-T cell proliferation, memory phenotype, and tumor infiltration.IL-2-9-21 (a cytokine cocktail) delays CAR-T depletion and enhances the anti-tumor activity of aAPC-CAR-T combination therapy.[28]To develop a GMP-compatible quadruplex aAPC expansion system for large-scale expansion of donor γδ T cells.Quadruplex aAPC combined with zol/IL-2 can efficiently amplify γδ T cells with memory phenotype and strong cytotoxicity, which is suitable for clinical application.[29]To develop K562-derived aAPC to replace Peripheral Blood Mononuclear Cells (PBMCs) for efficient amplification and function preservation of melanoma TIL.The aAPC-amplified TIL is comparable to PBMC in terms of efficiency, phenotype, function, and clonal diversity, supporting clinical-grade applications.[30]To construct K562-derived aAPC to optimize the in vitro expansion efficiency and function of CD19-specific T cells.CD19/4-1BBL/MICA co-expressed aAPC combined with IL-15/IL-2 can efficiently expand highly active CD19R+ T cells and support clinical transformation.[31]To develop functional aAPC to optimize in vitro expansion and gene transfer of anti-tumor T cells and to reduce the cost of CAR-T production.aAPCs effectively promote T cell expansion, reduce depletion, and support efficient gene transfer, making them a cost-effective alternative to traditional methods.[32]To address the immunosuppressive nature of K562 cells and to construct efficient aAPC to amplify CMV-specific T cells.Formaldehyde-fixed genetically engineered K562 cells can be used as “off-the-shelf” aAPC to efficiently induce CMV-specific CTLs, which has potential for clinical application.[33]To develop a highly efficient cell-based aAPC system to compare the effect of K562 and U937 in CD4+ T cell expansion.K32/CD3/28 aAPC achieves rapid long-term CD4+ T cell expansion through the synergistic action of co-stimulatory molecules (B7-H3) and cytokines (IL-15), which is superior to the bead-based system.[34]To compare the effects of CD137L and CD28-mediated aAPC on the in vitro expansion of mouse CD8 cells.To validate the role of CD137L-modified neuroblastoma vaccine in inducing anti-tumor immunity in vivo.CD137L (on aAPC) mediates superior costimulation vs anti-CD28, efficiently expands murine CD8+ cells, and induces DX5+ activated cell populations in vitro.In vivo CD137L vaccine enhances Th1-type antitumor immunity by activating CD8+DX5+ memory effector cells.[35]To develop a standardized aAPC platform for efficient amplification of TIL.To compare the amplification efficiency and phenotypic function difference between aAPC and traditional methods.aAPC realizes efficient amplification of TIL at a low ratio and has significant phenotypic advantages.Establishment of “off-the-shelf” platform to support clinical-grade TIL production.[36]Mouse fibroblast cell lineTo assess aAPC expressing HLA-A2.1 in activating and amplifying CMV-specific CTLs and to make a multifaceted comparison with autologous APCs.The aAPCs could efficiently amplify CMV-specific CTLs, were functionally equivalent to autologous APCs, and had the utility of obtaining clinical-grade specific CTLs with a single stimulation.[37]NIH-3T3 Cell LineTo investigate whether aAPC can activate and amplify HLA-A∗0201-restricted hTERT-specific CTLs, process hTERT antigens and present p540 and p865 epitopes to induce tumor-killing CTLs, and evaluate their clinical potential to solve the time-consuming problem of autologous APC preparation with traditional protocols.The aAPC can efficiently present hTERT-derived p540/p865 epitopes, can process endogenous antigens and present dominant epitopes, CTLs with tumor-killing ability, and high in vitro amplification efficiency with clinical potential.[38]To construct the aAPC expressing HLA class II, B7.1, ICAM-1, and LFA-3 to assess its ability to stimulate and expand specific memory CD4+ T cells in vitro and to compare it with natural APCs.To explore the ability of aAPC to process intact protein antigens and present epitopes to induce Th1 cell differentiation in primary T cell culture.aAPC efficiently presents antigen, is superior to B-EBV cells, and is more advantageous in memory CD4+ T cell expansion, providing a new tool for immunotherapy.[39]To construct an aAPC to amplify MART-1 specific CTLs with stem cell-like memory features in vitro for melanoma treatment.aAPC efficiently generates high-purity and highly functional CTLs, which possess strong cytotoxicity and TSCM phenotype, providing a new tool for adoptive immunotherapy.[40]To construct HLA universal aAPC and to evaluate its efficiency in inducing virus-specific T cells and epitope response profile.The aAPC effectively generates HLA-restricted T cells, amplifies dominant/subdominant epitope responses, and has "off-the-shelf" immunotherapeutic potential.[41]MDA-MB-231 Breast Cancer Cell LineTo develop a highly efficient aAPC system (Med-APC) to replace autologous DCs and to solve the problem of antigen-specific CD8+ T cell induction in vitro.To validate the ability of Med-APC to present viral/tumor antigens and to optimize the ubiquitination vector.Med-APC is equivalent to mature DCs and efficiently induces/expands functional CTLs.Ubiquitination vector enhances endogenous antigen presentation and activates T cells without exogenous peptides.[42]Human CD34+ Hematopoietic Progenitor CellsTo develop erythrocyte-derived aAPC to activate T cells and to induce anti-tumor immune responses through triple signaling.Engineered erythrocyte aAPC effectively promotes T cell expansion, memory formation and tumor control with a favorable safety profile, providing a new strategy for allogeneic “off-the-shelf” tumor immunotherapy.[43]Mature Mouse Red Blood cellsTo develop IL2-tethered RBC-based aAPCs (R-aAPC-IL2) to activate T cells and induce anti-tumor immune responses.R-aAPC-IL2 effectively promotes T cell proliferation, cytokine secretion and antigen-specific cancer cell killing, providing a new platform for cancer immunotherapy.[44]Peripheral Blood LymphocytesTo construct homologous targeted aAPC (LC-aAPC) based on peripheral blood lymphocytes.To mimic the distribution of natural APCs ligands to achieve efficient T cell activation and tumor immunotherapy in vivo.LC-aAPC retains the homing ability through lipid-DNA technology and can effectively activate antigen-specific T cells in vitro/in vivo; combined with anti-PD-1, it significantly inhibits tumor growth, providing a new strategy for personalized immunotherapy.[45] Summary of aAPCs based on modified cell lines with antigen-presenting functions for tumor immunotherapy. Another famous cell line is the NIH-3T3 mouse fibroblast cell line, which has also been studied by many researchers [37]. Similar to K562, NIH-3T3 cells are characterized by low immunogenicity and ease of manipulation. For example, Dupont and colleagues reported a study in Cancer Research in 2005. In this research, they transfected NIH-3T3 cells with HLA-A∗0201, CD80, and CD54 molecules to construct an aAPC. This aAPC, when expressing either hTERT-derived p540/p865 minigenes or full-length hTERT, efficiently stimulated tumoricidal hTERT-specific CTLs, with approximately 100 million such CTLs generated from 100 mL blood after five sequential stimulations (expansion of ∼1 × 105-fold) [38]. Since then, numerous studies have been conducted based on NIH-3T3 cells [[39], [40], [41]]. Additionally, researchers have explored other cell lines, such as the MDA-MB-231 breast cancer cell line [42] and human CD34+ hematopoietic progenitor cells [43]. What's more, a unique approach was presented by Sun et al., in 2017 [44], where they developed an aAPC by directly modifying the surface of mature mouse red blood cells. Through the biotin-avidin bridge, pre-formed pMHC-I, anti-CD28 and IL-2-PEG complexes were directly affixed onto the erythrocyte surface, offering a simpler preparation process compared to gene engineering and avoiding potential safety risks associated with the integration of foreign genes. However, this approach has shortcomings, including dependence on the coupling efficiency for surface molecule density and a tendency for molecules to detach, resulting in a limited signaling duration. Similarly, Sun and colleagues modified peripheral blood lymphocytes to create lymphocyte-based homologous targeting artificial antigen-presenting cells (LC-aAPCs), which were then re-infused to activate T cells in vivo [45]. Overall, modified cell lines offer a robust and scalable platform for ex vivo T cell expansion, with well-established protocols and the ability to customize co-stimulatory molecule expression. Their key limitations include the risk of foreign gene integration, the need for irradiation before in vivo administration to prevent proliferation, and potential immunogenicity of the scaffold cells, which restricts their in vivo applicability. Enhanced natural APCs In contrast to modifying non-immune cell lines, this approach seeks to enhance the intrinsic antigen-presenting capabilities of natural APCs—such as DCs—through physical or chemical interventions, thereby improving their stability and functional potency while preserving their natural machinery. This design approach, aimed at improving the functionality of natural APCs, has been explored less extensively compared to the modification of existing cell lines. One significant study by Lin et al., published in Advanced Materials in 2021, introduced a strategy for enhancing DCs by incubating them with hydrogel solutions. The hydrogel monomers permeated the cells, and UV light was used to create a three-dimensional hydrogel network inside the cells. This network provided mechanical support, significantly improving the stability and operability of the DCs without compromising their antigen-presenting functions. The key advantage of this approach is that it enhances both the stability and functionality of natural APCs while maintaining their ability to effectively present antigens, leading to a substantial increase in overall efficiency, as demonstrated by robust CFSE dilution and increased T cell yields in vitro [47]. Consequently, in the tumor model, the expanded T cells reduced tumor volume by 60% within 20 days and increased the survival rate of the mice to 80%, compared to 40% in the control group. This strategy offers the advantage of preserving the full repertoire of natural antigen-presenting molecules and avoiding genetic modification. However, its broader application is constrained by the limited availability of autologous APCs, batch-to-batch variability, and the technical complexity of the enhancement process, which together limit scalability for widespread clinical use. Particle-based aAPCs In contrast to cell-based aAPCs, particle-based aAPCs are composed of particles derived from natural or synthetic materials, thereby functioning as aAPCs and fulfilling a specific role. The research of this type of aAPCs is more focused on the innovation of material types and characteristics. According to the origin of their particles, the particle-based aAPCs can be divided into three classifications: natural particle-based aAPCs, synthetic particle-based aAPCs, and hybrid particle-based aAPCs. Each of these aAPCs can be classified into detailed groups, and they offer distinct advantages in enhancing T cell activation and therapeutic efficacy. The merits of each particle will be expounded upon in the subsequent section. Natural particle-based aAPCs Natural particle-based aAPCs mainly refer to various types of vesicles with antigen presentation function, such as exosomes. The source of the vesicle membrane can be various types of cells, among which DC membrane-based vesicle construction is a more common construction strategy. This category of aAPCs has a high degree of biocompatibility and frequently functions as a vehicle for various drugs, antibodies, and other therapeutic agents. Furthermore, it is distinguished by its low immunogenicity and toxicity, as well as a prolonged in vivo circulation. In the following categories, aAPCs are grouped according to the origin of the membranes. DCs derived vesicles The prevalence of DC-based studies can be attributed to the inherent functionality of DCs as antigen-presenting cells, thereby obviating the necessity for subsequent modification of antigen-presenting molecules on the vesicle surface. A study demonstrated in an in vitro assay that DC membrane-based extracellular vesicles effectively stimulated antigen-specific CD8+ T cells, resulting in a notable elevation of 2-fold TNFα, 1.8-fold IFNγ, and 1.7-fold CD107a compared to the inactivated group [48]. In a separate study, researchers designed a DC-based exosome with a diameter of 186 ± 7.6 nm and a zeta potential of −29.1 ± 2.2 mV. The DC2.4 cells were cultivated in a specialized medium and then researchers collected the supernatant. The exosome fraction was purified via differential ultracentrifugation to obtain the DC2.4-derived exosomes, which were denoted as “Exo”. Subsequently, tumor-specific neoantigens were loaded into Exo via electroporation, yielding “Exo-neoantigen” carriers. For example, Exo-M27/M30 was obtained by encapsulating M27/M30 peptides, which correspond to B16F10 melanoma, within exosomes. This exosome effectively curbed tumor development and metastasis by fully activating the immune system. In MC-38 and B16F10 models, the tumor volume of Exo-neoantigen group was reduced by about 50% compared to the liposome group.The 30-day tumor-free rate of the Exo-M27/M30 group reached 60%, while the control group had all the morbidity. In the lung metastasis model, the number of metastatic nodules in the Exo-M27/M30 group was reduced by 70% compared to the liposome group [49]. These findings suggest that DC-based vesicles have considerable potential for use in tumor immunotherapy and offer significant value for translational medicine applications. Macrophages derived vesicles Macrophage derived vesicles also possess distinct advantages. On one hand, they retain antigen-presenting capacity while exhibiting intrinsic tumor-targeting and tissue-infiltrating abilities, with high efficiency in phagocytosis and antigen capture. On the other hand, they can enhance immune activation and reshape the TME through M1 polarization. Additionally, they demonstrate excellent in vivo stability, prolonged circulation time, easy cell cultivation, and high extracellular vesicle production yield, making them highly suitable for clinical translation. In a separate study, Jiang and colleagues reprogrammed macrophages via the PIB transcription factor to construct engineered macrophage-derived nanovesicles, MAC-PNV, which possessed both DC-like antigen-presenting capacity and M1-type pro-inflammatory properties [50]. These nanovesicles elevated the CD8+ T cell activation ratio to 62.77% and increased proliferation efficiency by 2.23-fold. Meanwhile, they reprogrammed M2-type tumor-associated macrophages toward an M1 phenotype and boosted the secretion of pro-inflammatory cytokines including IFN-γ and TNF-α by 2.5–4-fold. As a single agent, MAC-PNV achieved a 77.00% tumor growth inhibition rate and extended the median survival of tumor-bearing mice from 21 days to 33 days. When combined with low-dose liposomal doxorubicin in B16-F10 tumors, the combination achieved a tumor inhibition rate of 78.65%, which was significantly higher than that of either agent alone, and the lung metastasis inhibition rate reached 77.20%. This strategy significantly enhanced anti-tumor and anti-metastatic efficacy while avoiding myelosuppression and organ toxicity associated with high-dose chemotherapy. Other cells derived vesicles Other cells derived vesicles as artificial antigen-presenting cells have shown different properties in tumor immunotherapy. AML cell-derived ECNV-αGC [51] has an average diameter of ∼359 nm and a surface zeta potential of −28.8 mV. It remains stable in serum for 11 days. By preserving the MHC-I-tumor antigen complex and loading αGC, ECNV-αGC directly activates iNKT cells and leukemia-specific CD8+T cells in a mouse model. It improves leukemia-free survival in the mouse model and can activate iNKT cells from human AML patients. Besides, tumor cell (EG7) derived EXOEG7 [52] contains OVA antigen and MHC-I, but lacks co-stimulatory molecules. Therefore, they rely on host DCs to stimulate CD8+T cells and their efficacy is weaker than DC-derived exosomes. In another research, IL-12-anchored renal cancer cell exosomes [53] were about 52 nm in diameter, carried G250 antigen, promoted T cell proliferation and IFN-γ release, and enhanced CTL toxicity. In another study [54], researchers treated aAPC with cytochalasin B, followed by vortexing and centrifugation. Subsequently, DNase1 was administered to avert the occurrence of aggregation, and the mixture was thoroughly washed with Phosphate Buffered Saline (PBS). Finally, this process yielded Antigen-Presenting Cytochalasin-B-Induced Membrane Vesicles (AP-CIMVs). AP-CIMVs were about 637 nm in diameter, expressed HLA and co-stimulatory molecules, and activated T cells when loaded with exogenous peptides but have limited endogenous antigen presentation. Overall, these vesicles are nanoscale and rely on MHC-I-antigen to activate CD8+ T cells, yet differ in DC dependence, modification strategies and endogenous antigen presentation, providing diverse directions for tumor immunotherapy. Compared with DCs derived vesicles, other cells derived vesicles are more widely available and their acquisition is thus relatively less difficult, such as the AML cell-derived ECNV [51], which can be prepared directly from the patient's own leukemia cells without additional DC cell culture steps, making it more suitable for personalized therapy. At the same time, their antigens are also more versatile and some of the vesicles are relatively simpler to prepare. However, the stability and homogeneity of some other cells-derived vesicles are relatively poor, and some of them are still dependent on host DCs for their function, limiting their functional independence. Synthetic particle-based aAPCs Synthetic particle-based aAPCs are aAPCs that are produced using physical and chemical techniques. The aAPCs obtained are unique due to the differences in the physical and chemical properties of the raw materials, as well as the differences in the synthesis strategies. The categorization of synthetic particulate-based aAPCs is proposed, which includes the following six groups: liposomes, PLGA, latex, magnetic beads, DNA-based aAPCs, and emerging micro-/nano-particles materials. Liposomes Liposomes stand out from other materials due to their phospholipid bilayer structure, which closely resembles that of natural cell membranes, offering excellent biocompatibility [55]. Moreover, liposomes can carry or adsorb various molecules on their surface, enabling the integration of multiple functions. By introducing pH-sensitive or temperature-sensitive phospholipids, liposomes can also achieve controlled release of therapeutic agents [56]. A study by Chen et al., published in Advanced Healthcare Materials in 2023, took advantage of liposomes’ multifunctional integration. They generated highly uniform, cell-sized (22.54 ± 1.39 μm) lipid bilayer vesicles (aAPC-BLMs) using a droplet microfluidic approach, demonstrating critical membrane fluidity (generalized polarization value = −0.597 ± 0.089). By coupling multiple antibodies to the surface of the liposome, they successfully created an aAPC that promoted efficient antigen-specific T cell activation, achieving a substantial 233-fold expansion of NY-ESO-1-specific CD8+ T cells in human peripheral blood mononuclear cells (PBMCs) at a low aAPC:T cell ratio of 1:17, without requiring repetitive boosting or additional soluble stimulants like cytokines [57]. Similarly, Giannoni and colleagues examined the effect of ligand clustering on T cell activation, using liposomes as a platform to couple relevant ligands and develop an aAPC [58]. Additionally, Zappasodi et al. employed liposomes enriched with GM1 as carriers, leveraging the fluidity of the phospholipid bilayer. And then they could better imitate the molecular clustering patterns seen in natural immune synapses. This approach enhanced the efficiency of T cell interactions [59]. Besides, lipid nanoparticles (LNPs) exhibit substantial promise in cancer immunotherapy, particularly when integrated with other therapeutic modalities. In 2024, Metzloff et al. reported in Advanced Materials the application of artificial LNPs (aLNPs) as a non-viral alternative to conventional CAR-T cell transduction. With a hydrodynamic diameter of 189.7 ± 97.3 nm, aLNPs can directly mimic the activation function of APCs by directly coupling CD3 and CD28 antibody fragments on the surface, eliminating the need to activate T cells with magnetic beads and subsequently remove the beads in the traditional method. In a mouse leukemia xenograft model, CAR-T cells generated by aLNPs reduced tumor load by 2.09-fold and prolonged survival by 6 days compared with the control group. After co-culturing with leukemia cells for 4 days, a tumor killing rate of 96.1% was still maintained at a 1:1 cell ratio, which demonstrated their long-lasting anti-tumor activity. This approach significantly shortens the time required for ex vivo T cell expansion while enhancing safety [60]. Overall, liposomes are a great material to build aAPCs. Actually, the exploration of materials for synthetic particle-based aAPC construction is not limited to liposomes alone. Current research on micro- and nano-particles primarily focuses on their physicochemical characterization and the selection of constituent materials (Table 3). Extensive studies have demonstrated that particle morphology, structure, and size critically influence the function of aAPCs [[61], [62], [63],94,96,97,100,[102], [103], [104], [105],108,109]. Among various material platforms, poly(lactic-co-glycolic acid) (PLGA) nanoparticles and latex microspheres have garnered particular attention. Table 3Summary of synthetic particle-based aAPCs for tumor immunotherapy.Table 3Particles materialsCharacterizationDesign approachPurposeEffectsRef.LipsomesAverage particle size: aAPC-shells precursor (DEDs): 22.36 ± 1.09 μmaAPC-BLMs precursor (SCMs): 22.54 ± 1.39 μmOn one hand, DEDs were directly conjugated with anti-CD28, anti-LFA-1, and MHCII-NYESO immunomolecules via biotin-streptavidin conjugation to generate aAPC-shells. On the other hand, DEDs were converted into SCMs in PBS before conjugation with the aforementioned molecules, yielding aAPC-BLMs.To address issues with natural APCs cultivation, including cumbersome procedures, high costs, and unstable quality, and the risk of malignant cloning in traditional cell lines.To validate the role of aAPC's cell size and membrane fluidity in T cell activation.To compare the differences in antigen-specific T cell activation between aAPC-shells and aAPC-BLMs.To explore the feasibility of preparing aAPCs using microfluidic technology.In PBMCs, aAPC-BLMs efficiently activate T cells with greater potency than aAPC-shells. Cellular size of aAPCs is a prerequisite for T cell activation, while membrane fluidity exerts a positive effect on T cell activation.[57]Average particle size of MDaAPC: approximately 100–1000 nmA basic carrier was constructed by assembling a lipid bilayer membrane from phosphatidylcholine and cholesterol, then co-incorporating GM1 ganglioside. Conventional aAPCs were then created by randomly inserting pMHC into this membrane. In contrast, MDaAPCs were constructed by pre-incubating biotinylated pMHC, biotinylated anti-CD28 antibody, biotinylated CTB, and NA. CTB then specifically bound to GM1 on the lipid membrane, forming ligand-enriched microdomains on the membrane.To construct MDaAPCs for simulating the discrete aggregation characteristics of T cell ligands on natural APCs surfaces.To validate the in vitro T cell activation efficiency of MDaAPCs by comparing them with conventional aAPCs, soluble MHC-peptide tetramers, and natural APCs.To investigate whether MDaAPCs can effectively trigger the physiological signaling pathways for T cell activation.To validate the feasibility of using MDaAPCs as tools for the in vitro regulation of antigen-specific T cells.MDaAPCs demonstrate significantly higher efficiency in activating T cells in vitro compared to conventional aAPCs and soluble MHC-peptide tetramers. MDaAPCs effectively promote the translocation of PKCθ, a key signaling molecule for T cell activation, from the cytoplasm to the cell membrane, thereby efficiently triggering the physiological signaling pathways involved in T cell activation. MDaAPCs have been demonstrated to serve as a viable tool for regulating antigen-specific T cells in vitro.[58]Average particle size of MDaAPC: approximately 100–1000 nmUsing GM1-enriched liposomes as a carrier, aAPC was constructed by anchoring anti-CD3, anti-CD28, anti-LFA-1, and neutravidin through the interaction between surface GM1 and biotinylated cholera toxin B.To overcome the limitations of natural APCs being time-consuming and existing commercial artificial systems exhibiting suboptimal T cell interactions.To compare the novel aAPC with commercial artificial systems for validating the high efficiency of the aAPC.To explore the potential of the aAPC for large-scale clinical application.Compared to the two commercial artificial systems, aAPC more efficiently expands T cells while maintaining higher T cell survival rates without increasing the proportion of regulatory T cells. Furthermore, by adjusting the antibody concentration during aAPC preparation or the ratio between aAPC and T cells, the potential for large-scale clinical application of aAPC has been demonstrated.[59]mal-LNPs:Average particle size: 115.5 ± 58.7 nmPolydispersity index (PDI): 0.259 ± 0.036aLNPs (1:1 CD3:CD28):Average particle size: 189.7 ± 97.3 nmPolydispersity index (PDI): 0.263 ± 0.036Using maleimide-modified mal-LNPs as precursors, the enzymatically cleaved and reduced CD3/CD28 antibody fragments were coupled to the surface of the particles through the thiol-maleimide reaction to construct aLNPs with both T cell activation (CD3/CD28 dual signaling) and mRNA transfection functions.To develop aLNP as an alternative to magnetic beads for one-step T cell activation and mRNA transfection, to optimize the vector design, and to validate the efficiency of CAR-T cell preparation and anti-tumor function in vitro.aLNPs can mimic the function of APC efficiently through the coupling of surface antibody fragments and simplify the process of CAR-T cell production. The CAR-T cells generated by the one-step method showed strong tumor-killing ability in vitro and have potential for clinical translation.[60]PLGASpherical aAPC:Diameter: 224 nmZeta potential: −2.9 ± 0.8 mVEllipsoidal aAPC:Aspect ratio: 2.8Radius of curvature of long axis: 1.14 μmZeta potential: −2.2 ± 0.2 mVThe aAPC was constructed by first making spherical nanoparticles with PLGA, then stretching them into ellipsoidal shape, and then modifying pMHC-dimers and anti-CD28 mAb on the surface of the particles using EDC/NHS chemical coupling method.To evaluate the effect of the shape of nano-ellipsoidal aAPC on T cell activation and in vivo distribution;To construct biodegradable nanoscale aAPC for in vivo immunotherapy.Ellipsoidal aAPCs significantly enhance in vitro T cell activation and in vivo T cell expansion by optimizing interfacial curvature and reducing non-specific uptake; biodegradable PLGA substrates combined with the ellipsoidal morphology enhance the pharmacokinetic properties and therapeutic potential of aAPCs.[61]Spherical diameter: 128 ± 5 nmProlate ellipsoidal long axis radius of curvature: 362 nmOblate ellipsoidal radius of curvature: 324 nmPLGA was used as the basis for three types of spherical, prolate ellipsoid, and oblate ellipsoid, which were coupled to MHC-antigen complex and anti-CD28 and combined with anti-PD-1 antibody for synergistic immunotherapy.To investigate the effect of particle shape on the activation of T cells by nano-aAPC and anti-tumor effect in vivo, and to develop degradable and efficient anisotropic nano-aAPC platforms.Long ellipsoidal nano-aAPC significantly enhanced T cell activation and anti-tumor efficacy by increasing T cell contact area and improving pharmacokinetics. In combination with anti-PD-1 therapy, long ellipsoidal aAPC effectively inhibited tumor growth and prolonged survival in mice, which was superior to conventional spherical nano-aAPC.[62]Spherical microparticles:number-weighted average diameter: 4.3 μmvolume-weighted average diameter: 6.7 μmBased on the film-stretching method, ellipsoidal PLGA microparticles with different aspect ratios were made, coupled with MHC-lg dimer and anti-CD28, respectively, while keeping the volume, protein density, etc. consistent in order to exclude interference, and only changing the shape.To investigate the effect of particle shape on the efficiency of CD8+ T cell activation by aAPC and to verify whether ellipsoidal aAPC enhances the immune response by increasing membrane contact.Ellipsoidal aAPC can increase the contact area to promote T cell binding and proliferation, and its activation efficiency in vitro is better than that of spherical particles. In vivo experiments further confirmed that ellipsoidal aAPC significantly inhibited tumor growth and prolonged the survival of mice, suggesting that shape is a key parameter in the design of aAPC.[63]Average hydrodynamic diameter of particles: 245±2 nmPLGA-based encapsulation of two cytokines, IL-2 and TGF-β, followed by modification of antibodies on the particle surface leading to specific binding and eventual release of cytokines in a paracrine form.To develop aAPC to induce human CD4+/CD8+ Tregs and to validate their in vitro/in vivo functions.aAPC effectively induced functional Tregs and alleviated autoimmune disease symptoms in humanized mice, demonstrating its immunotherapeutic potential.[64]Average diameter: 4.5 ± 1.2 μmZeta potential: 36.2 ± 5.6 mVPLGA-based, encapsulated IL-2 and anti-CTLA-4, surface-coupled H-2Kb/TRP2-Ig dimer and anti-CD28 antibody.To develop aAPC co-delivery of IL-2 and anti-CTLA-4 to enhance T cell immune response.PLGA-based aAPCs release IL-2 and anti-CTLA-4 via paracrine, which synergistically promote the expansion of antigen-specific T cells and effectively inhibit tumor growth.[65]Diameter: mean diameter 4.5 ± 1.5 μm, 82.5% 3-7 μmZeta potential: 38.6 ± 5.1 mV after PEI modification, returned to neutral after coupling proteinIL-2, IL-15, CCL21, anti-CTLA-4, and anti-PD-1 were encapsulated inside PLGA, which was PEI-treated with surface covalently coupled H-2Kb/TRP2180-188-Ig dimer, H-2Db/gp10025-33-Ig dimer, anti-CD28, anti-4-1BB, anti-CD2 and CD47-Fc.To develop a multimolecular synergistic aAPC (MaAPC) to enhance tumor antigen-specific T cell expansion and anti-tumor immune response.MaAPC, which carries 11 immune molecules, significantly enhances T cell activation in vitro and tumor suppression in vivo, which is superior to the traditional aAPCs carrying fewer immune molecules (usually 2-3 kinds), and provides a new strategy for individualized tumor immunotherapy.[66]Spherical PLGA-NPs:Average diameter: 214.3 ± 26.6 nmZeta potential: - 2.3 ± 0.2 mV, modified by PEI 20.7 ± 2.4 mVEllipsoidal PLGA-NPs:Zeta potential: 21.7 ± 3.1 mV, PEGylated 22.3 ± 2.2 mVEllipsoidal stretching of PLGA spherical nanoparticles, PEGylated modification and CD47-Fc coupling, and synergistic coupling of H-2Kb/TRP2 antigenic peptide and anti-CD28, to construct ellipsoidal nanoparticles of aAPC, which combine anti-phagocytosis and efficient activation of T cells.To construct “tealth” nano-ellipsoidal aAPC with anti-phagocytosis and efficient activation of T cells to enhance anti-tumor immune response.Ellipsoidal stretching, PEGylation, and CD47 conjugation synergistically reduce phagocytosis, promote antigen-specific T cell expansion, and effectively inhibit tumor growth.[67]Median diameter of plain microspheres: 7.78 μmMedian diameter of surface-modified microspheres: 7.06 μmThe aAPC was constructed by encapsulating IL-2 with PLGA, then modifying the microsphere surface with an affinity-palmitate conjugate, and then using biotin-affinity-specific conjugation to couple a biotinylated HLA-A2/influenza peptide complex and an anti-CD28 antibody to the microsphere surface.To develop PLGA-based degradable aAPC for IL-2 slow release and dual signaling modification for efficient stimulation of influenza peptide-specific CTLs.The aAPC can significantly promote the proliferation of CTLs through the sustained release of IL-2 and surface dual signaling, which is better than autologous APC, and verifies its potential for in vitro expansion of T cells.[68]PLGA particles:Diameter: 1.1 ± 0.3 μmZeta potential: −5 ± 2 mV (at pH 7.4 and pH 6.0)PLGA/PBAE particles:Diameter: 1.3 ± 0.3 μmZeta potential: −3 ± 1 mV (at pH 7.4), 24 ± 1 mV (at pH 6.0)Degradable PLGA and PBAE were mixed at a 3:1 wt ratio. The positive charge and immunostimulatory properties of PBAE were harnessed to enhance surface protein (e.g., MHC-Ig dimer and anti-CD28 antibody) coupling efficiency and T cell binding ability, followed by EDC/NHS chemical conjugation to construct aAPCs.To develop the aAPC, a novel PLGA/PBAE hybrid polymer, to enhance T cell activation and anti-tumor effects in vivo and to circumvent the limitations of in vitro cell manipulation.PLGA/PBAE aAPC effectively promotes T cell expansion and tumor suppression in vivo through enhanced protein presentation and T cell binding, and the combination therapy significantly improves the efficacy of melanoma models.[69]LatexLatex bead diameter: 5 μmHLA-A2/pIL-13Ra2345-354tetramers, anti-CD28 antibody, and CD83 molecules were sequentially encapsulated on the surface of 5-μm sulfated polystyrene latex beads to construct aAPC.To induce IL-13Ra2-specific CTLs using HLA tetramer-modified aAPC and to verify its anti-glioma activity.The aAPCs efficiently stimulated specific CTLs and significantly killed antigen-positive tumor cells in vitro, which has potential for clinical application.[70]Latex bead size: “cell-sized”The aAPC was constructed by sequentially encapsulating streptococcal affinity factor, anti-CD28 antibody, CD54 molecule and biotinylated HLA-A2-pLMP2 monomer on polystyrene sulfate latex beads as a carrier.To develop HLA tetramer-modified aAPC to induce LMP2-specific CTL in vitro.The aAPC efficiently amplifies antigen-specific CTLs with HLA restriction and tumor cell killing ability.[71]Latex bead diameter: 5 μm5 μm polystyrene sulfate latex beads were used as a carrier, sequentially coated with streptavidin, anti-CD28 antibody, CD83 molecule, and finally combined with biotinylated HLA-A2/pIL-13Ra2345-354 tetramer complexes to construct aAPC.To induce allogeneic-restricted CTLs targeting HLA-A2+ malignant glioma cells.Successful preparation of specific CTLs by aAPC confirmed their killing activity and HLA-I dependence on HLA-A2+/IL-13Ra2+ cells.[72]Latex bead size: “cell-sized”Cell-sized latex microbeads were used as carriers, and aAPCs were constructed by coating H-2Kb-Ig/pTRP2 complex, anti-CD28 antibody, 4-1BBL, and CD83 at precise concentrations on the surface of the beads.To develop a bead-based aAPC to efficiently amplify melanoma-specific CTLs and to verify its in vitro killing activity and in vivo anti-tumor effect.The aAPC successfully amplified high-purity pTRP2-specific CTLs, and the induced CTLs significantly inhibited the growth of melanoma in vitro and in vivo, confirming the practical value of aAPC.[73]Latex bead size: “cell-sized”Cell-sized latex microbeads were used as carriers, and the H-2Kb-Ig/pTRP2 complex, anti-CD28 antibody, 4-1BBL, and CD83 were coated on the bead surface. The beads were rotationally incubated at 4 °C for 24 h in 0.1 M borate buffer, washed, and then re-incubated for an additional 24 h. Finally, aAPCs were constructed by loading with pTRP2 peptides at a final concentration of 60 μg/mL.To amplify melanoma antigen-specific allo-restricted CTLs in vitro and assess the anti-tumor effect in vivo.aAPC effectively amplified allo-restricted CTLs and showed targeted anti-melanoma activity both in vivo and ex vivo, with immunotherapeutic potential.[74]Latex bead diameter: 5.3 μmThe aAPC was constructed using 5.3-μm polystyrene sulfate latex beads, sequentially coated with streptavidin-allophycocyanin, recombinant human CD80/Fc chimera, recombinant human CD54, 1% human albumin, and finally biotinylated HLA-A2/HA-1 or HLA-A2/HA-2 complexes.To verify whether aAPC can replace natural APCs to efficiently stimulate and enrich mHag-specific CTLs.aAPC can effectively activate CTL clones and polyclonal lines, maintain the function, and enrichment efficiency is comparable to that of DC, so it has potential for clinical application.[75]Latex bead diameter: 5 μmUsing 5-μm sulfated polystyrene latex microspheres as a substrate, anti-His6 monoclonal antibody and anti-mouse CD28 monoclonal antibody were co-coated on the surface of the microspheres in the ratio of 1:1, and then, through the specific binding effect of the anti-His6 antibody, the HLA-A2/tumor antigen peptides (HER2/neu369-377 or MART127-35) tetramer coupled to the microspheres, thus constructing the aAPC.To validate the ability of latex microsphere-based aAPC to induce tumor antigen-specific CTL responses in naïve mice.The ability of aAPC to effectively promote CTL proliferation and specific killing through co-delivery of antigens and co-stimulatory signals supports its use in active immunotherapy of tumors.[76]Latex bead diameter: 5 μmThe aAPC was constructed by coupling H-2Kb/TRP2 tetramer to the bead surface via anti-His6 antibody on 5 μm diameter sulfated polystyrene latex beads, while anti-CD28 and anti-4-1BB antibodies were coated in a 1:1:1 ratio.To validate the in vivo induction of tumor-specific CTL responses and anti-tumor effects of latex bead-based aAPC in natural T cell banks.aAPC can effectively activate CTL and inhibit tumor growth by delivering tumor antigen and co-stimulatory signals, which provides a new strategy for active immunotherapy of tumor.[77]Latex bead diameter: 5 μmRecombinant I-Ad/LACK dimer (an immunodominant peptide containing the Leishmania protozoa antigen LACK bound to the MHC class II molecule I-Ad) and anti-CD28 co-stimulatory antibody were encapsulated on the surface of the microbeads at optimized concentrations (40 μg/mL for I-Ad/LACK and 2 μg/mL for anti-CD28) using 5-μm polystyrene sulfate latex microbeads as a carrier to construct the aAPC.To develop CD4+ T cell-targeted synthetic aAPC and to validate its ability to activate initial T cells and anti-tumor immunity.Synthetic aAPC activates CD4+ T cells through MHC-II/CD28 signaling and induces protective anti-tumor responses in vivo, providing a new strategy for cell-free vaccines.[78]Magnetic BeadsDiameter: 4.5 μmConjugate HLA-A2 Ig and anti-CD28 human monoclonal antibody to the surface of magnetic beads, then incubate them in antigen peptide solution for 24 h to construct the aAPC. Store the resulting aAPC in the original peptide solution at 4 °C.To validate the in vivo function of tumor-specific CTLs amplified in vitro using aAPCs, including their in vivo survival capacity and tumor-targeting ability.To evaluate the antitumor efficacy of these CTLs in both tumor prevention and tumor treatment models, comparing them with conventional approaches to clarify the application potential of aAPCs.aAPC can efficiently expand Mart-1-specific CTLs and effectively suppress tumor growth, with therapeutic efficacy comparable to that of DC-expanded CTLs. It holds broad application prospects in cancer adoptive immunotherapy.[79]Magnetic bead diameter: 2.8 μmHER2 was biotinylated, loaded onto magnetic beads, and blocked with biotinylated BSA, thereby constructing the aAPC.To address limitations of traditional target cell lines such as unstable antigen density and poor physiological relevance.To validate whether aaTPs can specifically and dose-dependently induce activation and effector functions of anti-HER2 CAR-T cells.To demonstrate the potential of aaTPs as a standardized tool for CAR-T development and clinical translation.aAPC exhibits controllable HER2 density, excellent biocompatibility, and antigen-dependent modulation of CAR-T cell activity. It can specifically induce CAR-T activation and cytokine secretion, thereby validating next-generation CAR-T mechanisms.[80]Diameter of nano-aAPC: 80-100 nmUsing magnetic beads as carriers, humanized HLA-Ig dimers preloaded with melanoma-associated tumor antigen peptides were surface-conjugated to these beads. Simultaneously, anti-CD28 antibodies were also conjugated, thereby constructing aAPCs.To develop a highly efficient and rapid T cell expansion platform.To ensure the superior characteristics of expanded T cells.To validate the clinical utility of aAPC, and compare it with conventional methods.aAPC rapidly and efficiently expands tumor antigen-specific CD8+ T cells, which exhibit a stem cell-like memory phenotype, high functional activity, and long telomeres. aAPC demonstrates efficacy in both treatment-naive and PD-1 antibody–treated patients with stage IV melanoma and outperforms conventional methods.[81]Mouse aAPC diameter: Approximately 80 nmHuman aAPC diameter: Approximately 140 nmMix Tz-labeled nanoparticles with TCO-labeled pMHC-I dimers and anti-CD28 antibodies, incubate overnight at 4 °C, then purify to obtain aAPCs.To achieve efficient in vitro expansion of antigen-specific mouse/human TRM-like CD8+ T cells, address the weak homing capacity of TRM cells following intravenous delivery.To validate the local/systemic antitumor immune responses of hydrogel-delivered TRM-like cells,To investigate synergistic effects with anti-PD-1 checkpoint inhibitors, and demonstrate clinical translational potential.aAPC efficiently induces antigen-specific TRM-like CD8+ T cells. Hydrogels enhance TRM-like cell delivery efficiency. TRM-EH exhibits potent antitumor effects with favorable synergistic activity when combined with anti-PD-1, demonstrating promising clinical translation potential.[82]Magnetic microsphere diameter: 4.5 μmaAPCs were constructed by chemically conjugating HLA tetramers and anti-CD28 antibodies to magnetic beads via different chemical coupling methods.To construct aAPCs based on HLA tetramers to activate human CD4+ T cells in an antigen-specific manner, addressing the challenges of DC vaccine preparation and T cell expansion in vitro.All kinds of aAPCs could activate human CD4+ T cells in an antigen-specific manner, but activation efficiency varied significantly; indirectly conjugated aAPCs (particularly those mediated by anti-Class II MHC antibodies) emerged as the optimal tool for efficient antigen-specific activation.[83]NAFor different antigens, soluble MHC-Ig fusion proteins are first conjugated with specific peptide fragments to form pepMHC-Ig complexes. These pepMHC-Ig complexes and B7.1-Ig molecules are then covalently coupled to magnetic beads, thereby constructing the aAPC.To develop an in vivo deliverable aAPC system to activate adoptively transferred low-affinity tumor-specific CTLs.To validate the enhanced anti-tumor activity of aAPC on low-affinity CTLs in clinical models.To elucidate the mechanism by which aAPC activates low-affinity CTLs.aAPC reverses the in vivo anti-tumor ineffectiveness of low-affinity CTLs, induces their in vivo proliferation, enhances their in vitro function, and demonstrates considerable efficacy in animal models.[84]Bead size range: 50 nm, 300 nm, 600 nm, 4500 nmaAPCs were constructed by conjugating chimeric pMHC-Ig carrying SIY model antigens and anti-CD28 monoclonal antibodies to the surface of SPIONs at a 1:1 M ratio.To investigate the influence of aAPC physical parameters (particle size, concentration, etc.) on CD8+ T cell activation quantitatively.To verify whether matching aAPC size with T cell surface TCR nanoclusters and pMHC patches could enhance activation efficiency.To explore strategies to overcome low activation efficiency of small-sized aAPCs.To compare with endogenous APCs to advance aAPC application in adoptive cell therapy.aAPC activation of CD8+ T cells exhibits “particle size dependency.” 50 nm aAPCs require “saturation dose” for efficient activation, and increasing ligand density alone cannot resolve activation defects. aAPCs with particle sizes above 300 nm show no significant differences in activation characteristics. Magnetic artificial clustering overcomes activation deficiencies in 50 nm aAPCs at “sub-saturation doses.”[85]Elastic modulus: 14.6 kPaSPION diameter: 100 nmUsing microfluidics, SPIONs were encapsulated synchronously during droplet formation to construct magnetic alginate spheres. Stability was further enhanced through dual cross-linking. Subsequently, anti-CD3 and anti-CD28 amino groups were conjugated to carboxyl groups on the microsphere surface. After purification, the aAPC was obtained.To investigate the synergistic effect of oscillatory mechanical force and aAPC physical parameters on T cell activation.To determine whether “absolute antigen quantity” or “antigen density” is more critical for T cell activation.To compare the advantages and disadvantages of homemade versus commercial aAPCs.To optimize aAPC signal intensity to achieve efficient expansion, stable maintenance, and functional enhancement of induced regulatory T cells (iTregs).Larger aAPC particle size and higher absolute antibody content correlate with stronger T cell proliferation. This indicates that “absolute antigen quantity” is the key determinant of T cell activation.Oscillating mechanical force synergistically promotes selective expansion of CD8+ T cells with large-particle aAPCs.Homemade aAPCs outperform commercial aAPCs.The particle size, antibody load, and mechanical force of aAPCs enable precise regulation of T cell activation or iTreg differentiation, offering a novel strategy for cellular immunotherapy.[86]DNAEdge length: about 120 nmUsing triangular DNA origami as a 2D scaffold, it was biotinylated and then biotinylated pMHC was anchored at predetermined sites on the scaffold via biotin-streptavidin (SA) specific binding. Subsequently, DNA-linked aCD28 was site-specifically conjugated to the three scaffold vertices. Following stepwise ultrafiltration purification, four distinct aAPC-Tn variants were obtained, differing in pMHC copy number and spacing.To develop DNA origami-based aAPCs that precisely regulate surface pMHC and aCD28 to optimize CD8+ T cell activation.To validate the specific anti-tumor killing efficacy of aAPCs in vivo and in vitro, providing a novel tool for adoptive T cell therapy.aAPC effectively activated CD8+ T cells in vitro. Higher pMHC density resulted in longer binding residence time between aAPC and TCR. It demonstrated potent tumor-killing effects both in vivo and in vitro.[87]Average particle size: 1.83 ± 0.17 μmUsing RCA technology, DNA microflowers (DM) with a multi-lobed porous structure were prepared. The surface was then modified with PDA coating, followed by conjugation of anti-CD3 and anti-CD28 antibodies. Finally, the porous structure was loaded with hIL-2 to construct aAPC.To address the issues of high resource consumption in traditional in vitro T cell stimulation and low contact efficiency between existing carriers and T cells.To evaluate its in vitro T cell activation efficacy and in vivo immunotherapeutic effects in suppressing primary/distant tumors and preventing tumor recurrence.DM-AI significantly induced the expansion of CD8+ T cells while markedly reducing the proportion of Tregs, demonstrating favorable in vitro antitumor effects. In vivo, it enriched at tumor sites, completely inhibiting distant tumor growth and suppressing primary tumors.[88]Gold nanoparticlesAuNPs:Core Diameter 1: 15 nm, Total Diameter: 50 nmCore Diameter 2: 30 nm, Total Diameter: 90 nmCore Diameter 3: 50 nm, Total Diameter: 140 nmSpiky AuNPs were used as carriers to synthesize 15/30/50 nm core diameter spiky structures using seed-mediated method to match the spacing of T cell signaling molecules (35-70 nm) to promote multivalent binding; αCD3 and αCD28 antibodies were coupled on the surface via DNA-streptavidin bridging system and induced T cell autocrine cytokines.After screening, the 50 nm core size and the ligand ratio of αCD3:αCD28 = 3:1 were determined, and the ideal aAPC was finally obtained.To develop spiked gold nanoparticles as aAPC to optimize in vitro T cell activation and expansion.The spiked structure enhances T cell activation and reduces depletion through high ligand density and multivalent binding, which is superior to existing systems for CAR-T cell therapy.[89]PolydopaminePDA MVs:Diameter: approximately 1 μmPolydopamine microbubbles (PDA MVs) with a cellular flexible vesicle structure (∼1 μm in diameter) were used as carriers, and anti-CD3 and anti-CD28 antibodies were covalently modified on their surfaces using the Sulfo-SMCC coupling method to construct aAPC.To develop a nucleus-inhibiting microemulsion method to synthesize cell-like PDA MVs and to validate the ability of PDA MVs to act as aAPC to activate T cells.Flexible PDA MVs significantly enhance T cell activation efficiency by mimicking natural cellular structures, while the nucleation inhibition strategy provides a new approach for the controlled synthesis of functionalized microvesicles.[90]alginateMicrogels:Diameter: 77 ± 2 μm, 72 ± 2 μm after LbL coatingThe aAPC was constructed by wrapping a polyelectrolyte layer around the microgel surface via layer-by-layer (LbL) technology, enabling efficient coupling of stimulatory ligands (e.g., αCD3/αCD28 antibodies and MHC-I antigens). This strategy allows precise modulation of ligand density, ratio, and spatial distribution to regulate T cell expansion and phenotype.To develop surface-functionalized microgels as tunable aAPCs to regulate T cell expansion and phenotype, and to analyze the mechanism of surface ligands and mechanical properties of microgels on T cell activation.Microgel aAPCs efficiently activate mouse/human T cells and support polyclonal and antigen-specific amplification. At the same time, the density and ratio of surface ligands and microgel stiffness can precisely regulate T cell expansion efficiency and functional phenotype.[91]Nanoparticle diameter: 550 ± 40 nmZeta potential: −26.9 ± 0.5 mVThe aAPC was constructed by preparing nanoparticles from alginate using the oil-in-water nanoemulsion method, and then coupling anti-CD3 (or KbSIY peptide - MHC-I) with anti-CD28 antibody to the surface of the particles by the EDC/NHS chemical method.To develop alginate-based aAPC to modulate oxidative stress and phenotypic differentiation of CD8+ T cells and to enhance the memory properties and in vivo efficacy of adoptive T cell therapy.Alginate-based aAPC can reduce ROS and promote memory-related transcription factors to achieve efficient expansion of CD8+ T cells, and significantly enhance their in vivo persistence and anti-tumor ability.[92]Chip channel aperture size: 50 μmArtificial antigen-presenting gel droplets (AAPGD) diameter:Water-oil phase flow rate ratio: 1:1 →AAPGD diameter: 50 μmWater-oil phase flow rate ratio: 1:2 →AAPGD diameter: 30 μmModify Anti-CD3 and Anti-CD28 on the surface of alginate gel via carbodiimide method, co-encapsulate cytokines (IL-2, IL-15, IL-21) and CMP within the alginate gel, then form microdroplets with fluorocarbon oil using a droplet microfluidic device (both aqueous and oil phases at 20 μL/min). Trigger Ca2+ release with acetic acid to complete gelation, thereby obtaining uniform aAPC.To address the issues of low efficiency of T cell expansion in vitro and weak tumor infiltration capacity in vivo in adoptive cell therapy (ACT).In vitro, T cell proliferation rates increased, with low-differentiated T cells accounting for 82% of the total, and increased secretion of tumor-killing-related factors.In vivo, in a melanoma mouse model, tumor volume decreased to 6.7 mm3 (15 days), with a survival rate of 100% (35 days), and complete tumor regression was achieved when used in combination with anti-PD-L1.[93]paramagnetic iron-dextranIron dextran nanoparticles aAPC: average diameter 50-100 nmQuantum dot nanoparticles aAPC: diameter ∼30 nmBased on two nanoplatforms, iron dextran paramagnetic particles and avidin-coated quantum dot nanocrystals, aAPCs were constructed by coupling MHC-peptide complexes with co-stimulatory molecules (e.g., B7.1-Ig or anti-CD28 antibody) to the surface of the particles in a 1:1 ratio via chemical coupling or a biotin-avidin system.To develop nanoscale aAPC and to verify its in vitro T cell activation and in vivo anti-tumor ability.Nanoscale aAPC induced antigen-specific T cell proliferation efficiently and inhibited tumor growth in vivo, with better particle distribution and stronger immune activation than micron-sized particles.[94]Diameter: 50-100 nmThe aAPC was constructed by co-coupling MHC-peptide dimer and anti-CD28 antibody to paramagnetic iron-dextran nanoparticles as a carrier.To develop nano-aAPC (nano-artificial antigen-presenting cell) platform for efficient enrichment and rapid expansion of tumor-specific T cells.The nano-aAPC-mediated "Enrichment + Expansion" (E + E) strategy significantly improves the expansion efficiency of tumor-specific T cells, which is validated in vivo and ex vivo to respond to shared antigens and novel epitopes, providing a new tool for individualized immunotherapy.[95]dextranmolecular weight of the aAPS (700–900 kDa)Using dextran as a scaffold, aAPCs were constructed by first aminating and biotinylating it, then coupling biotinylated multimerized peptide-MHC complexes (e.g., SIINFEKL-β2m-H2-Kb) and B7-1 co-stimulatory ligands to the dextran backbone through streptavidin.To develop dextran-based aAPCs to study their ability to activate, differentiate and rescue CD8+ T cells.The aAPCs have potential for cancer immunotherapy by efficiently inducing T cell activation, proliferation and CTL differentiation through dual signaling (pMHC + B7-1) and repairing hyporesponsive/unresponsive T cells.[96]streptavidin-conjugated dextran molecular weight: 500 kDaThe aAPC was constructed by modular assembly of biotinylated pMHC, co-stimulatory molecules (e.g., B7.2), and cytokines (e.g., IL-2, IL-21) via biotin-streptavidin interaction modularization using streptavidin-coupled dextran as a flexible skeleton to form an Ag-scaffold.To develop dextran-based Ag-scaffolds for selective expansion of highly functional antigen-specific T cells for ACT.Ag-scaffolds efficiently amplify T cells with self-renewal, low depletion, and strong killing ability, and the multi-target strategy is applicable to melanoma patients.[97]polystyreneStreptavidin-coated polystyrene particles: approx. 5.6 μm in diameterThe aAPC was constructed by loading biotinylated HLA-A∗0201-ELAGIGILTV peptide monomers onto streptavidin-coated polystyrene beads (via biotin-streptavidin interaction) as the substrate, and co-coating with costimulatory antibodies at an optimal ratio of 3:1 for anti-4-1BB to anti-CD28.To optimize the ratio of aAPC co-stimulatory signals (CD28 and 4-1BB) to enhance the efficiency of antigen-specific CTL amplification.The 3:1 ratio is the best synergy for efficient generation of effector memory CTLs, which validates the feasibility and superiority of the synthesized aAPC platform.[98]Microsphere diameter: 5.6 μmUsing streptavidin-coated polystyrene microspheres as a carrier, aAPC was constructed by co-coating biotinylated HLA-A∗0201/HA-1H peptide complexes (at high density [10 nM] or low density [100 pM]) with biotinylated anti-CD28 antibody, taking advantage of the microspheres' biotin-binding capacity (0.064 μg biotin-FITC per mg microsphere).To validate the feasibility of aAPC to efficiently induce HA-1-specific CD8+ effector memory T cells in vitro.aAPC successfully induced Tc1-type effector memory T cells with high cytotoxicity, which has potential for clinical application.[99]semi-flexible polyisocyanide peptidesSemi-flexible filamentary polymers:Diameter: 5-10 nmLength: 150-200 nmDegree of polymerization (DP): 1633Statistical spacing of azide groups: 10 nmAntibody-functionalized synthetic DCs (sDCs), or aAPCs, were constructed by using a semi-flexible filamentous poly(isocyano peptide) polymer as a scaffold, introducing an azide moiety through random copolymerization, and using SPAAC to click chemically couple to streptavidin (SAv), which was then conjugated to a biotinylated anti-CD3 antibody.To develop aAPC to mimic the multivalent receptor interactions and dynamic activation capabilities of natural DCs and to verify the effect of structural flexibility and multivalency on T cell activation efficiency.aAPCs activate T cells more efficiently at low concentrations, inducing faster and longer-lasting immune responses. Structural flexibility and multivalent ligand synergism are key factors in enhancing T cell activation over rigid spheres or free antibodies.[100]Average spacing of azide groups: approx. 10 nmAverage polymer length: 150-200 nmUsing a semi-flexible poly(isocyano peptide) as the polymer backbone, the polymer was formed by copolymerizing methoxy-terminated and azide-functionalized isocyanate monomers (with azide groups statistically spaced). Then, 3−5 streptavidin molecules were coupled to the polymer via SPAAC, followed by sequential binding of biotinylated anti-CD3 (αCD3) and anti-CD28 (αCD28) antibodies, resulting in the construction of the aAPC.To develop bifunctional semi-flexible polymer-based sDC (aAPC) to optimize T cell activation and induce efficient anti-tumor immune responses.aAPC significantly enhances T cell activation, proliferation and generation of multifunctional killer T cells, and inhibits regulatory T cells, providing a new strategy for tumor immunotherapy.[101]Length: about 400 nmWidth/Thickness: 2 nmUsing semiflexible polyisocyanopeptides (PICs) as a backbone, aAPCs were constructed by multivalent anchoring of peptide-major histocompatibility complexes (pMHCs) with co-stimulatory molecules (e.g., αCD28) or cytokines (e.g., IL-2) to the polymer surface via strain-promoted azide-alkyne click chemistry.To develop nanoscale aAPC to directly activate antigen-specific T cells in vivo to fight tumors.The aAPC effectively promotes T cell expansion and inhibits tumor growth and metastasis, and it is more effective in combination with immune checkpoint blockade.[102]poly(ethylene glycol)-block-poly(D,L-lactide)(PEG-PDLLA) copolymerSmall-size spherical (SmS): about 160 nm in diameter.large-size spherical (LgS): about 300 nm in diameter.small-size tubular (SmT): about 200 nm × 70 nm in size.large-size tubular (LgT): about 2 μm × 50 nm in size.Using PEG-PDLLA as the material, polymer vesicles with four morphologies, spherical and tubular, of different sizes were prepared by the solvent exchange method combined with extrusion and osmotic pressure-induced shape transformation techniques. aAPC libraries were constructed by covalently coupling αCD3 and αCD28 antibodies to the surface of the vesicles at high, medium, and low densities using the SPAAC click reaction.To investigate the effects of aAPC morphology (spherical/tubular), size (large/small) and ligand density on T cell activation.High ligand density and large size enhanced activation, tubular structure was more effective at low ligand density, and bifunctionalization was superior to monofunctionalization.[103]biotin-PEG-PHPMA(-SH)-PDMAMean particle size of NP-OVA: 47 nmMean particle size of NP-HPPH: 58 nmMean particle size of NP-DOX: 52 nmMean particle size of NP-OVA after 24 h in pH 5.0 environment: 1800 nmThe redox-sensitive copolymer biotin-PEG-PHPMA (-SH)-PDMA was synthesized by RAFT polymerization and then self-assembled into a nanovesicle (BCNV), whose surface was modified by the biotin-streptavidin system with SIINFEKL peptide-loaded MHC I monomers and αCD28, and the internal water nucleus was loaded with IL-2, thus constructing the aAPC.To develop size-variable aAPCs combined with nanotechnology to pre-activate CD8+ T cells and enhance anti-tumor immunity in vivo.Redox-responsive aAPC were transformed to micrometer scale within the tumor, and combined with nanoplatforms effectively activated T cells and significantly inhibited tumor growth.[104]ZnO nanowire4 μm nanowire bending modulus: 440 kPa9 μm nanowire bending modulus: 88 kPa29 μm nanowire bending modulus: 20 kPaThe aAPC was constructed by preparing vertical ZnO nanowires by chemical vapor deposition, functionalizing anti-CD3 antibody on the surface to provide chemical stimulation, and modulating the topology and bending modulus using different lengths of nanowires to match the cumulative shear modulus of the nanowire forests to the elasticity range of myeloid and DCs in vivo.To investigate the mechanism of synergistic regulation of T cell activation by physical properties (elasticity/topology) of nanowires and chemical signaling of anti-CD3 antibody.Nanowire elasticity (regulated by length) and antibody signaling synergistically enhance T cell signaling, degranulation, and cytotoxicity, and mechanical stimulation acts through an actin-dependent pathway.[105]poly(N-isopropylacrylamide) (PNIPAM)SPM:Hydrodynamic diameter: about 1 μmZetapotential: +32.7 mVDPM:Hydrodynamic diameter: about 1 μmZetapotential: +29.1 mVUsing poly N-isopropylacrylamide (PNIPAM) as the material, aAPC was constructed by introducing the hydrophilic monomer MPC to form a dendritic structure similar to that of natural DCs, and then utilizing the high affinity of Protein A for IgG to adsorb it on the surface of positively charged microspheres, which in turn directed the immobilization of CD3ε and CD28 antibodies.To construct efficient aAPC using bionic dendritic microspheres to enhance T cell activation and tumor immunotherapy.The dendritic morphology significantly enhances the T cell expansion capacity of aAPC and effectively inhibits tumor growth in vivo.[106]Composite particle systems (cell-mimetic silica particles + lipid bilayers)Average diameter of erythrocyte templated particles: 4.5 μmDiameter of spherical microspheres: 8 μmAverage diameter of HeLa cell templated particles: 16 μmUsing HeLa cells, human erythrocytes, or commercial smooth spherical silica microspheres as templates, silica particles were synthesized by dilute silicic acid solutions and calcined to remove the biotemplate. Small unilamellar vesicles (SUVs) containing biotin were then fused with the particles at a 1:1 mass ratio to form a supported lipid bilayer, and biotinylated anti-CD3/CD28 antibodies were coupled via biotin-avidin chemistry, thus constructing the aAPC.To investigate the effects of aAPC shape, size, membrane fluidity, and antibody density on T cell expansion and differentiation.High antibody density, fluid membrane and non-spherical particles promote balanced T cell expansion and maintain undifferentiated phenotype, which is superior to commercial activation beads.[107] Summary of synthetic particle-based aAPCs for tumor immunotherapy. PLGA PLGA has been widely employed to activate and expand T cells [[64], [65], [66], [67], [68]]. Most studies adopt a similar strategy in constructing aAPCs, wherein functional molecules are conjugated onto the PLGA particle surface, although the specific molecular components differ across applications. Song et al. constructed aAPCs primarily by coupling H-2Kb/TRP2180-188-Ig dimers and anti-CD28 antibodies onto the PLGA surface, in addition to altering the PLGA morphology. Meanwhile, they further developed PEGylated nanoellipsoidal aAPCs with additional CD47-Fc coupling to enhance their phagocytosis-inhibiting capacity [67]. Beyond surface conjugation of functional molecules, some designs focus on encapsulating immunomodulatory agents within PLGA to achieve sustained release. In other cases, PLGA is combined with other materials. For example, Rhodes et al. reported in 2021 that hybrid particles composed of PLGA and poly(beta-amino-ester) (PBAE) improve both stability and targeting efficiency, while maintaining biocompatibility [69]. Latex Latex microspheres have also been extensively utilized [[70], [71], [72], [73], [74], [75], [76], [77], [78]]. Jiang et al. demonstrated the conjugation of HLA-peptide complexes and costimulatory molecules onto latex microspheres for the induction of CTLs [70]. Interestingly, both PLGA- and latex-based aAPCs follow a similar modular design principle: antigenic and costimulatory signals are presented on the particle surface to mimic the dual-signal activation mechanism of natural APCs. However, differences in material properties dictate distinct application scenarios. Latex microspheres, being non-degradable and highly stable, with well-established surface modification techniques, are typically suited for in vitro T cell expansion. By contrast, PLGA offers superior biocompatibility, tunable degradation rates and particle sizes, and dual loading capabilities (surface and core), making it more applicable for in vivo immune modulation. Magnetic beads Magnetic beads are uniquely characterized by their magnetic responsiveness, which enables targeted delivery, modulation of intercellular interactions, and efficient retrieval and separation. As early as 2009, magnetic beads conjugated with HLA-Ig and antigenic peptides were employed by Durai et al. to achieve efficient in vitro expansion of CTLs [79]. Since then, magnetic beads have been widely investigated for the construction of aAPCs [[80], [81], [82], [83], [84]]. Despite addressing varied scientific questions, most of these studies adopt comparable design principles for aAPC development. Notably, a growing number of studies have been inspired by the magnetic properties of beads, integrating these features with other material platforms to enhance aAPCs functionality. In a 2017 study published in Nano Letters, superparamagnetic iron oxide nanoparticles (SPIONs), which are nanoscale magnetic beads, were conjugated with pMHC-Ig and anti-CD28 antibodies to activate T cells. Quantitative analysis revealed a striking size-dependence: aAPCs larger than 300 nm (e.g., 300, 600, 4500 nm) induced robust T cell proliferation (∼12-fold expansion), while 50 nm aAPCs at equivalent pMHC dose (2 pM) elicited significantly weaker responses (∼5-fold expansion). It was emphasized that particle size must exceed a critical threshold to facilitate effective engagement with multiple TCR nanoclusters [85]. In a related study, also in Nano Letters, SPIONs were embedded within alginate hydrogels to explore the synergistic enhancement of T cell activation via mechanical oscillations and scaffold flexibility [86]. Both studies focus on the application of SPIONs, addressing scientific challenges from distinct perspectives—particle size and material flexibility. These investigations exemplify the strategic integration of magnetic responsiveness with the structural and mechanical advantages of complementary nanomaterials, representing a promising direction for advancing aAPC engineering. DNA-based aAPCs In addition to excellent biocompatibility and multifunctional integration, DNA-based aAPCs offer the unique advantage of nanoscale precision in molecular arrangement. Leveraging complementary base pairing, DNA nanostructures can be assembled into highly customizable two- or three-dimensional architectures, allowing precise spatial control over the orientation, density, and positioning of key molecules such as pMHC and costimulatory ligands (e.g., anti-CD28 antibodies). This meticulous organization mimics the immunological synapse of natural APCs and significantly enhances T cell activation. Consequently, DNA nanotechnology represents a promising and innovative platform for aAPCs development. However, the construction of DNA-based aAPCs is technically demanding—requiring meticulous geometric design and sequence matching in both 2D (two-dimensional) and 3D (three-dimensional) configurations. In addition, structural characterization is often complex and time-intensive, limiting widespread adoption and exploration. To date, two primary strategies have been explored: DNA origami and DNA microflowers (Table 4). In 2022, Sun et al. reported in Science Advances the successful construction of DNA origami-based aAPCs capable of robustly activating T cells [87]. These triangular DNA nanostructures enabled precise spatial organization of activating ligands (anti-CD28 at vertices, pMHC at edges) with nanoscale control over density (9-36 copies) and inter-ligand spacing (20-60 nm). Crucially, single-particle tracking revealed that increasing pMHC density prolonged the TCR-pMHC binding dwell time from 9.9 s to 12.1 s, correlating with enhanced T cell activation markers. In vivo, these constructs suppressed tumor growth and prolonged mouse survival without observed toxicity. This work shows DNA origami's potential as a promising tool for immune modulation, though further optimization is needed for broader application. In a distinct approach, Le et al. employed rolling circle amplification (RCA) to construct DNA microflowers from specific DNA templates and primers in the presence of ligase and DNA polymerase [88]. These were coated with dopamine to anchor anti-CD3 and anti-CD28 antibodies, and loaded with human IL-2 to complete the aAPC platform. While functionally promising, DNA microflowers—like DNA origami—pose substantial fabrication challenges and require continued refinement. Table 4Comparison of DNA-based aAPCs for tumor immunotherapy.Table 4DNA origami [87]DM-AI [88]Materials1.2D nanoscaffolds2.Anchored ligands via biotin-streptavidin (SA) coupling3.Precisely structurally controllable (vertex and edge preset sites)1.Prepared by rolled-circle amplification (RCA)2.Three-dimensional porous flower-like structure (diameter ∼1.83 μm)3.Coated with polydopamine (PDA)Ligand modification1.Vertex-anchored anti-CD28 antibody2.Edge-anchored pMHC3.Density adjustable1.Covalent binding of anti-CD3 and anti-CD28 antibodies after surface PDA coating.2.Porous structure loaded with hIL-2Structural characteristicsTwo-dimensional planar structure with nanoscale precise ligand localization and optimized spatial distribution of ligands by adjusting pMHC densityThree-dimensional porous, petal-like pleated surface increases T cell contact area; internal cavities can be loaded with cytokines for sustained releaseSignal combinationsDual signaling system:1.pMHC (TCR antigen-specific recognition, first signal)2.anti-CD28 (co-stimulation, second signal).Three signaling systems:1.anti-CD3 (TCR activation, first signal)2.anti-CD28 (co-stimulation, second signal)3.hIL-2 (cytokine, third signal).T cell mechanism of action1.High-density pMHC prolongs pMHC-TCR binding time and enhances calcium signaling and cytokine secretion2.Ligand spatial arrangement promotes TCR clustering and enhances signaling intensity1.Three-dimensional pleated surface increases membrane contact area and promotes antibody binding to TCR2.Sustained release of hIL-2 promotes CD8+ T cell proliferation and inhibits Treg differentiationin vitro effect1.Proportion of CD69+ T cells: increased with the increase of pMHC density2.Cell proliferation rate: aAPC-T12 group was significantly higher than the mixed ligand group1.Proliferation index of CD8+ T cells: 4.1 times higher than that of PS-A2.Secretion of IFN-γ: increased 1.8 times3.Tumor cell killing rate: DM-AI group (72%) vs PS-A group (10%).in vivo effect1.Adoptive immunotherapy significantly inhibited the growth of B16-OVA tumors and reduced tumor volume in the aAPC-T12+T cell group compared to the control group2.Histology showed increased tumor cell apoptosis and prolonged survival in mice1.Local injection inhibited primary tumor growth and induced immune rejection of distant tumors2.Increased proportion of CD8+ T cells and decreased Treg in the TMEBiocompatibilityDNA origami scaffolds with high stability and low cytotoxicityDNA backbone with PDA coating, degradable, no apparent toxicity in vivo Comparison of DNA-based aAPCs for tumor immunotherapy. 1.2D nanoscaffolds 2D nanoscaffolds 2.Anchored ligands via biotin-streptavidin (SA) coupling Anchored ligands via biotin-streptavidin (SA) coupling 3.Precisely structurally controllable (vertex and edge preset sites) Precisely structurally controllable (vertex and edge preset sites) 1.Prepared by rolled-circle amplification (RCA) Prepared by rolled-circle amplification (RCA) 2.Three-dimensional porous flower-like structure (diameter ∼1.83 μm) Three-dimensional porous flower-like structure (diameter ∼1.83 μm) 3.Coated with polydopamine (PDA) Coated with polydopamine (PDA) 1.Vertex-anchored anti-CD28 antibody Vertex-anchored anti-CD28 antibody 2.Edge-anchored pMHC Edge-anchored pMHC 3.Density adjustable Density adjustable 1.Covalent binding of anti-CD3 and anti-CD28 antibodies after surface PDA coating. Covalent binding of anti-CD3 and anti-CD28 antibodies after surface PDA coating. 2.Porous structure loaded with hIL-2 Porous structure loaded with hIL-2 1.pMHC (TCR antigen-specific recognition, first signal) pMHC (TCR antigen-specific recognition, first signal) 2.anti-CD28 (co-stimulation, second signal). anti-CD28 (co-stimulation, second signal). 1.anti-CD3 (TCR activation, first signal) anti-CD3 (TCR activation, first signal) 2.anti-CD28 (co-stimulation, second signal) anti-CD28 (co-stimulation, second signal) 3.hIL-2 (cytokine, third signal). hIL-2 (cytokine, third signal). 1.High-density pMHC prolongs pMHC-TCR binding time and enhances calcium signaling and cytokine secretion High-density pMHC prolongs pMHC-TCR binding time and enhances calcium signaling and cytokine secretion 2.Ligand spatial arrangement promotes TCR clustering and enhances signaling intensity Ligand spatial arrangement promotes TCR clustering and enhances signaling intensity 1.Three-dimensional pleated surface increases membrane contact area and promotes antibody binding to TCR Three-dimensional pleated surface increases membrane contact area and promotes antibody binding to TCR 2.Sustained release of hIL-2 promotes CD8+ T cell proliferation and inhibits Treg differentiation Sustained release of hIL-2 promotes CD8+ T cell proliferation and inhibits Treg differentiation 1.Proportion of CD69+ T cells: increased with the increase of pMHC density Proportion of CD69+ T cells: increased with the increase of pMHC density 2.Cell proliferation rate: aAPC-T12 group was significantly higher than the mixed ligand group Cell proliferation rate: aAPC-T12 group was significantly higher than the mixed ligand group 1.Proliferation index of CD8+ T cells: 4.1 times higher than that of PS-A Proliferation index of CD8+ T cells: 4.1 times higher than that of PS-A 2.Secretion of IFN-γ: increased 1.8 times Secretion of IFN-γ: increased 1.8 times 3.Tumor cell killing rate: DM-AI group (72%) vs PS-A group (10%). Tumor cell killing rate: DM-AI group (72%) vs PS-A group (10%). 1.Adoptive immunotherapy significantly inhibited the growth of B16-OVA tumors and reduced tumor volume in the aAPC-T12+T cell group compared to the control group Adoptive immunotherapy significantly inhibited the growth of B16-OVA tumors and reduced tumor volume in the aAPC-T12+T cell group compared to the control group 2.Histology showed increased tumor cell apoptosis and prolonged survival in mice Histology showed increased tumor cell apoptosis and prolonged survival in mice 1.Local injection inhibited primary tumor growth and induced immune rejection of distant tumors Local injection inhibited primary tumor growth and induced immune rejection of distant tumors 2.Increased proportion of CD8+ T cells and decreased Treg in the TME Increased proportion of CD8+ T cells and decreased Treg in the TME Emerging micro-/nanoparticle materials A range of emerging materials has also been explored for aAPCs development. Gold nanoparticles, as investigated by Esmaeili et al., offer excellent biocompatibility but are costly and synthetically demanding [89]. Polydopamine microbubbles studied by Dong et al. provide mechanical flexibility, enhancing signal delivery. However, particle uniformity is suboptimal and degradation is slow, raising potential safety concerns [90]. By contrast, particles composed solely of alginate are cost-effective and easily manufactured, yet offer limited functionality. Consequently, existing research has primarily focused on diversifying their functional applications [[91], [92], [93]]. Paramagnetic iron-dextran nanoparticles supports efficient magnetic enrichment and precise cell separation with good biocompatibility, but it is non-degradable and reliant on external magnetic fields, complicating implementation [95]. Other materials under investigation include polystyrene microspheres [98,99], semi-flexible polyisocyanide peptides [101], and poly(N-isopropylacrylamide) (PNIPAM) [106]. Composite particle systems have attracted increasing interest. For example, Olden et al. presented silica particles coated with lipid bilayers and functionalized with surface antibodies to elicit aAPC-like activity [107]. Hybrid particle-based aAPCs Hybrid particle-based aAPCs are typically fabricated by coating nanoparticles with various types of cell membranes. Among these, DC membranes are most frequently employed, owing to their distinctive dendritic morphology, enhanced antigen uptake and presentation capabilities, widespread tissue distribution, and superior immunoregulatory functions compared to other cell types [[110], [111], [112], [113]]. Consequently, a substantial amount of research has focused on DC membrane-coated nanoparticles as a foundational platform for biomimetic aAPCs development. DC membrane-coated nanoparticles Earlier, in 2020, Cheng et al. described in Advanced Science the creation of "mini DCs" via the physical extrusion of native DC membranes onto PLGA nanoparticles, achieving successful application in ovarian cancer immunotherapy [114]. Characterization revealed these biomimetic nanoparticles possessed a core-shell structure with a uniform diameter of 160-170 nm (<200 nm), a size favorable for efficient trafficking to lymph nodes. Critically, these mini DCs demonstrated superior T cell activation compared to native DCs, inducing a threefold increase in CD69+-activated CD8+ T cells in vitro. Furthermore, in a murine ovarian cancer model, prophylactic vaccination with mini DCs led to a substantial tenfold reduction in abdominal metastatic nodules compared to the PBS control group, showcasing robust anti-tumor efficacy. Meanwhile, using click chemistry and metabolic glycoengineering, Xiao et al. modified DC membranes in a study published in Nano Letters in 2021. The modified membranes were then used to coat PLGA nanoparticles, a process that generated aAPCs with strong T cell activation efficiency [115]. These studies utilize unmodified or chemically engineered DC membranes for nanoparticle encapsulation and consistently demonstrate favorable immunotherapeutic outcomes in vitro and in vivo. Generally, these findings demonstrate the high degree of customizability and functional utility of DC membranes in the design of biomimetic aAPCs. Macrophages membrane-coated nanoparticles Macrophage membrane-coated nanoparticles represent another promising type of hybrid aAPC system. Similar to DC membranes, macrophage membranes inherit natural targeting, homing, and immune escape abilities, enabling prolonged blood circulation and enhanced accumulation in lymphoid organs or tumor tissues. Meanwhile, macrophage membrane-coated nanoparticles can effectively present antigens and provide costimulatory signals for T cell activation. Such biomimetic platforms combine the structural advantages of synthetic nanoparticles with the physiological functions of macrophage membranes, making them highly suitable for in vivo cancer vaccine and immunotherapy applications. Recent studies further demonstrated that macrophage membrane-coated nanoparticles possess exceptional ability to penetrate the blood–brain barrier (BBB), achieving 180.26-fold higher brain tumor accumulation compared with free drugs and 3.85-fold higher than uncoated PLGA nanoparticles. Engineered macrophage membranes with enhanced PD-1 expression can specifically block the PD-1/PD-L1 immune checkpoint pathway, which significantly promotes CD8+ CTL infiltration by 2.22-fold and increases the secretion of antitumor cytokines such as IFN-γ and TNF-α, thereby effectively reversing immunosuppression in the TME [116]. Nanoparticles coated with other engineered cell membranes Some biomimetic aAPCs derived from other various cell types are frequently designed to emulate the immunological functions of DCs. In 2023, Sun and his colleages modified 4T1 tumor cells to stably express pMHCs, co-stimulatory molecules and crucially, an anti-LAG3 antibody on the membranes, generating “super artificial DCs” (saDCs) [117]. The membranes of these engineered cells are subsequently employed to coat photosensitizer-loaded nanoparticles (specifically, nanoparticles loaded with an AIE photosensitizer, Fs, exhibiting NIR-II emission at 1012 nm), yielding a multifunctional nanoplatform that integrates immunotherapy with photodynamic therapy, demonstrating enhanced therapeutic efficacy. In other approaches, tumor cell membranes are directly utilized. For instance, Lu and co-workers apply membranes derived from B16-F10 murine melanoma cells to encapsulate CD-MnOx nanoparticles, resulting in the formation of CD-MnOx@CM [118]. To summarize, compared with other aAPCs modalities, hybrid particle-based aAPCs exhibit several distinctive advantages. These include the intrinsic biocompatibility and genetic modifiability of biological membranes, particularly those sourced from DCs. More critically, by coupling immune-responsive membranes—enriched with molecules such as MHCs and co-stimulatory ligands—with nanoparticles capable of therapeutic payload delivery, a synergistic system is established. This configuration enables both antigen-specific immune activation and precise, controlled drug release. Such an integrated strategy provides a unified platform for targeted antigen presentation, T cell stimulation (mediated by the membrane's immunological components), and the delivery of immune adjuvants or chemotherapeutic agents (mediated by the nanoparticle core). In the TME, this dual mechanism improves therapeutic effects in two ways. First, pMHCs and co-stimulatory signals initiate specific T cell responses. Second, the release of agents like imiquimod or IL-2 further amplifies immune activation or mitigates immunosuppression. Obviously, biomimetic aAPCs are positioned as an advanced iteration in the evolution of aAPC technologies. By leveraging the respective strengths of cell-based and material-based systems while addressing their inherent limitations, they represent a strategic convergence and a significant progression in the design of next-generation immunotherapeutics. Compared with natural and synthetic particle-based aAPCs, hybrid particle-based aAPCs show unique advantages in biomimetic immune evasion, systemic circulation stability, and synergistic drug delivery. Their core–shell structure enables efficient targeting to lymphoid organs and tumor tissues, making them particularly suitable for in vivo antitumor applications and personalized cancer vaccines. These features distinguish hybrid aAPCs as a highly promising platform for clinical translation. aAPC platform system Although aAPCs platform systems constitute a relatively novel design paradigm, their development is currently constrained by complex fabrication procedures. In contrast to the preceding two categories of aAPCs, which primarily involve nanoscale or microscale modifications to individual cells or particles, platform systems are assembled from millimeter-to centimeter-scale three-dimensional porous biomaterials, forming a macroscopic scaffold for immune modulation. These constructs are engineered to recapitulate natural APCs’ functional hallmarks. They work by spatially organizing multiple T cell activation cues within or on the scaffold matrix. These signals include TCR stimulation, co-stimulation, and cytokine release. This facilitates the formation of localized immune microenvironments conducive to robust T cell activation, proliferation, and functional differentiation (Table 5). Table 5Summary of aAPCs platform systems for tumor immunotherapy.Table 5MaterialsCharacterizationDesign approachPurposeEffectsRef.HA-PEGDA hydrogelElastic modulus: 0.2-3 kPaLigand density: 4 μg/mL for anti-CD3/CD28 (corresponding to about 500 nm spacing)Construction of stiffness-tunable hydrogels with thiolated hyaluronic acid (HA) and polyethylene glycol diacrylate (PEGDA) coupled with anti-CD3/CD28 dual-stimulation signaling to mimic the extracellular matrix microenvironmentTo mimic the extracellular matrix microenvironment to optimize T cell expansion efficiency and functional phenotype in vitro by modulating hydrogel stiffness and coupling anti-CD3/CD28 dual-stimulation signals.aTM promotes rapid proliferation of CD8+ T cells in soft matrix (0.5 kPa), induces memory phenotype, and enhances CD3 clustering and functional molecule co-expression through mechanotransduction.In a mouse melanoma model, aTM-stimulated antigen-specific T cells resulted in reduced tumor size and improved survival rate.[119]Diameter:50-300 μm storage modulus G′:0.5-10 kPa, optimal at about 2 kPa for T cell proliferation.Thiolated hyaluronic acid (HA) and polyethylene glycol diacrylate (PEGDA) were used to construct hydrogels with tunable stiffness, and then the hydrogels were fabricated into microgel particles on which peptide-loaded MHC complexes, anti-CD28 antibody, and IL-2/anti-IL-2 antibody complexes were anchored to mimic the natural lymph node microenvironmentTo mimic the natural lymph node microenvironment to achieve direct activation and efficient expansion of antigen-specific CD8+ T cells in vivo, avoiding the limitations of in vitro T cell manipulation in traditional immunotherapy.Rapid proliferation of CD8+ T cells in vitroRapid proliferation of OVA/SIY-specific T cellsGood in vivo targeted killing effect and long duration of activitySignificantly delayed the growth of B16-OVA tumors and prolonged survival of mice[120]alginate polymers exosome-cross-linked gelChimeric exosome size distribution: 40-160 nmInactivated GL261 tumor cells and BMDCs were induced to fuse, forming dendritic cell-tumor hybrid cells, which then secreted DT-Exos. Ac4ManNAz-mediated metabolic sugar engineering was used to edit the surface of DT-Exos to obtain N3-DT-Exos, which were then cross-linked with DBCO-bALG to form aAPC.To address the issues of weak T cell response and short duration of action in postoperative immunotherapy for GBM.The bALG@DT-Exo gel significantly promoted the proliferation and activation of CD8+ T cells in vitro, and remodeled the microenvironment in vivo, resulting in increased CD8+ T cell infiltration and reduced Tregs. The gel almost completely cleared residual lesions, significantly prolonged the median survival time of mice, and significantly reduced tumor invasiveness.[121]MSR-SLB ScaffoldMSR:Length:70 μm ± 31.7 μm/70 μmDiameter:5.1 μm ± 1.2 μm/4.5 μmPore diameter:5.34 nm/10.9 nmUsing MSR as a three-dimensional scaffold loaded with soluble cytokines and coated with liposomes containing biotinylated lipids to form SLBs, TCR-stimulated ligands and co-stimulated ligands were anchored through the biotin-streptavidin system, which mimicked the three-signal activation pattern of natural antigen-presenting cells and the dynamic microenvironment of the immune synapse.To mimic natural APCs' three-signal activation to achieve efficient T cell expansion and enrichment for adoptive cell therapy.APC-ms enabled excellent polyclonal expansion of primary T cells, achieved specific enrichment of rare T cell subsets, and effectively supported CAR-T cell expansion with clear translational promise for adoptive cell therapy.[122] [123]MSR:Length: 83.0 ± 32.3 μm (range 18.9-213.2 μm)Width: 14.6 ± 3.30 μm (range 4.5-23.0 μm)Using MSR as a three-dimensional scaffold loaded with soluble cytokines and coated with liposomes containing biotinylated lipids to form SLBs, TCR-stimulated ligands and co-stimulated ligands were anchored through the biotin-streptavidin system, which mimicked the three-signal activation pattern of natural antigen-presenting cells and the dynamic microenvironment of the immune synapse.To address the issue of limited efficacy in CAR-T cell therapy due to insufficient T cell persistence and aggravated exhaustion.At the cellular level, the T cell Enhancing Scaffolds (TES) enhance the proliferation capacity, persistence, and cytotoxicity of CAR-T cells, reduce exhaustion, and induce an effector memory phenotype. In animal models, a single TES injection significantly extended the survival period of lymphoma mice and reduced tumor burden even at non-curative CAR-T doses.[124]scaffold based on hydrogel of oxidized alginate-Calcium sulfateWhen the efficiency of T cell activation is maximum: Elastic modulus:50 kPaUsing an alginate-calcium sulfate hydrogel as a three-dimensional porous matrix, the scaffold is loaded with PLGA nanoparticles for sustained release of TGFβ inhibitors to suppress local Treg cells, integrated with mesoporous silica particles for slow release of IL-2 and CCL21 chemokines to recruit effector T cells, and surface-conjugated with anti-CD3/CD28 antibodies while tuning the elastic modulus to 50 kPa to potentiate T cell activation.To synergistically relieve the tumor-local immunosuppressive microenvironment, to recruit and activate effector T cells, and to induce an anti-tumor immune response while avoiding the toxicity risks of systemic immunotherapy.Remodeled the local immune microenvironment and promoted the induction of a pro-inflammatory immune microenvironment. Effectively inhibited tumor growth and circumvented systemic toxicity in tumor-bearing mice.[125]alginateAlginate scaffold thickness: 2 mmUsing porous silica microspheres as carriers, their cores can encapsulate and slowly release bioactive forms of IL-15 superagonist. At the same time, the microsphere surfaces are coated with a lipid bilayer and conjugated with anti-CD3, anti-CD28, and anti-CD137 antibodies. These microspheres are then embedded into the pores of GFOGER-modified alginate scaffolds to construct aAPCs.To address the issues of low efficiency in delivering T cells to tumor sites and insufficient proliferation of T cells in the immunosuppressive TME during adoptive T cell therapy for solid tumors, and to design a bioactive polymer scaffold to achieve the delivery, proliferation, and dispersion of tumor-reactive T cells for the treatment of unresectable or incompletely resected tumors.In a mouse breast cancer resection model, the scaffold effectively supported the presence of tumor-targeted T cells at the resection site and in associated lymph nodes, reducing tumor recurrence compared to traditional delivery methods. None of the mice receiving T cells via the scaffold exhibited tumor recurrence; In a multifocal ovarian cancer model, polymer-delivered T cells induced tumor regression, whereas injected tumor-reactive lymphocytes had virtually no therapeutic effect.[126]polycaprolactone microfibersMicrofibers: Diameter 3.4 ± 1.6 μm, length 55.5 ± 23.8 μmUsing polycaprolactone microfibers as a scaffold, dendritic cell membranes were coated to mimic their physicochemical properties. Through copper-free click reactions, anti-CD3 and anti-CD28 antibodies were immobilized on the surface to provide signal 1 and signal 2 for T cell activation, while adsorbing releasable IL-2 to provide signal 3, thereby constructing aAPC.To address the challenges of limited availability, functional impairment, and complex cultivation of autologous APCs in ex vivo T cell expansion for adoptive cell therapy, and to develop a scaffold system mimicking DCs to achieve efficient, scalable expansion of T cells with high functional activity.Promotes stronger polyclonal expansion of primary human T cells; at 14 days, the expansion fold of IL-2/Ab/DC@MF is 7 times that of Dynabead, and it is more biased toward CD8+ T cells;Antigen-specific T cell expansion was achieved via MHC on DC membranes, with the expanded T cells exhibiting target cell-specific killing capacity. Approximately 97% of the expanded T cells could be harvested via density gradient centrifugation, providing a scalable, modular, and customizable platform for adoptive cell therapy.[127]Alginate-PEG hydrogelAverage pore size: approx. 150 μmAnti-CD3/CD28 was conjugated to microspheres, which were then encapsulated in the pores of a macroporous hydrogel composed of Alg-N3 and DBCO-functionalized 8-arm PEG crosslinks, thereby constructing aAPC.To address the issues of time-consuming and costly in vitro expansion of existing CAR-T therapies, as well as the susceptibility of T cells to death or dysfunction after infusion, and to improve anti-tumor efficacy.Significantly promotes the expansion of CD8+ T cells in vitro, with normal function. Remains stable in vivo for several weeks, enhancing systemic CD8+ T cell response. In a preventive model, 80% of mice showed no tumor progression, demonstrating good antitumor efficacy.[128]Type I bovine collagenScaffold diameter: 21 mmUsing porous type I bovine collagen sponges as carriers, the scaffolds were functionalized with CCL21 to efficiently recruit host T cells, while lentiviral vectors encoding tumor-specific chimeric antigen receptors were co-encapsulated. Stimulatory anti-CD3/CD28 antibodies were added to construct aAPCs.To develop a scaffold to address the issues of time-consuming processes, high costs, low patient accessibility, and limited efficacy in solid tumors associated with existing TIL therapy and CAR-T therapy in vitro culture.In a mouse breast cancer model, 8 days after scaffold implantation, the proportion of CAR+ T cells at the tumor site reached 39.2 ± 19.5%, significantly higher than in the spleen. Scaffold therapy significantly inhibited tumor growth and extended the median survival time of mice. Additionally, this scaffold can be stored at 2–8 °C after large-scale preparation, making it suitable for clinical application.[129] Summary of aAPCs platform systems for tumor immunotherapy. Characteristics of aAPCs platform systems include a macroscopically visible 3D architecture that enables cellular infiltration and sustained signal propagation, along with the use of biocompatible materials for stable loading and controlled, spatiotemporal release of bioactive signals. These systems simultaneously permit ex vivo expansion of antigen-specific T cells and in vivo initiation of immune responses, providing a standardized and customizable “reactive base station” suitable for applications such as cancer immunotherapy. Commonly employed materials include hydrogels and biodegradable scaffolds. Hydrogel-based strategies can be broadly divided into two categories. One approach involves the direct fabrication of artificial T cell-stimulating matrices (aTMs). For example, Hickey and colleagues crosslinked thiolated hyaluronic acid (HA) with polyethylene glycol diacrylate (PEGDA), yielding a hydrogel with tunable mechanical stiffness (ranging from 0.2 to 3 kPa, optimized for T cell expansion below 1 kPa). The resulting matrix was functionalized with anti-CD3 or pMHC, along with anti-CD28 antibodies, to enable T cell activation [119]. Crucially, adoptive transfer of T cells expanded on this soft (0.5 kPa) aTM significantly suppressed tumor growth and improved survival (66% at day 40) in a murine melanoma model compared to traditional methods [119]. An alternative strategy focuses on engineering artificial lymph nodes (aLNs). Livingston and collaborators applied a similar HA-PEGDA hydrogel system, functionalizing the surface with pMHC, anti-CD28, and anti-CD3 antibodies, and IL-2/anti-IL-2 complexes. Microgel particles were subsequently generated through a sieving process [120]. The aTM serves as a customizable ex vivo 3D microenvironment for T cell activation and expansion, whereas aLN microgels, upon injection, aggregate into porous hydrogel scaffolds that mimic native lymphoid tissue microenvironment, supporting in situ T cell priming. In a separate study, researchers utilized inactivated GL261 tumor cells and BMDCs to formulate bALG@ DT-Exo as aLNs, which directly activated tumor-infiltrating T cells, significantly promoting CD8+ T cell proliferation (Ki67+ proportion reaching 31%) and activation (CD25+ proportion 14.0%) in vitro [121]. In vivo, the microenvironment underwent significant remodeling, leading to the near-complete elimination of postoperative residual lesions. This resulted in a notable extension of the median survival time of mice to 59 days, in comparison to the 36 days observed in the control group treated with cGAMP. Furthermore, magnetic resonance imaging (MRI) and tissue staining revealed a marked reduction in tumor invasiveness. Biodegradable scaffolds are distinguished by their capacity for sustained cytokine release and structural support. Their 3D architecture facilitates cellular colonization, while the integrated release systems maintain prolonged immune stimulation. These scaffolds provide mechanical integrity, undergo self-degradation, and exhibit excellent biocompatibility. Zhang and colleagues reported an aAPC-mimetic scaffold (APC-ms) in Nature Protocols, Nature Biotechnology and Nature Biomedical Engineering, which exhibited superior performance in polyclonal T cell expansion and selective enrichment of antigen-specific T cells [[122], [123], [124]]. Furthermore, this scaffold can also achieve local reactivation of CAR-T cells, significantly enhancing the antitumor activity of CAR-T cells [124]. These effects are attributed to the synergistic properties of the scaffold's architecture and its presentation of T cell stimuli combined with sustained release of cytokines such as IL-2. Further advancement has been achieved by Majedi and colleagues, who integrated hydrogel and biodegradable scaffold technologies [125]. In their design, alginate hydrogels were crosslinked with calcium ions to form porous, degradable matrices. The mechanical stiffness of the scaffolds was deliberately tuned based on Ca2+ concentration, with an optimal Young's modulus of ∼50 kPa identified to maximize T cell activation. The surface was covalently functionalized with anti-CD3/CD28 antibodies to initiate T cell activation, while the interior was loaded with nanoparticles carrying TGF-β inhibitors, the chemokine CCL21, and IL-2-releasing microparticles. This hybrid system leverages the structural integrity and degradability of scaffolds alongside the biocompatibility and signal-releasing properties of hydrogels, yielding an immune-activating microenvironment characterized by synchronized degradation and therapeutic signal delivery. This localized immunomodulation strategy demonstrated robust antitumor efficacy in vivo. In the aggressive 4T1 breast cancer model, the functionalized scaffolds significantly suppressed primary tumor growth in four out of five treated mice, whereas all control mice exhibited tumor growth. Furthermore, treated mice exhibited a significant survival advantage and showed complete elimination of metastases in the brain and axillary lymph nodes, which were universally present in untreated controls. In the B16-F10 melanoma model, mice receiving the full immunoactive scaffold demonstrated 80% long-term survival. In addition, several other materials and approaches were also used in the construction and design of the aAPC platform system [126]. In 2023, Kim et al. reported in Bioactive Materials the development of DCs membrane-coated polycaprolactone (PCL) microfibers (MFs) that function as artificial aAPCs. By mimicking natural DCs through surface presentation of CD3/CD28 ligands and IL-2, these scaffolds achieved superior polyclonal T cell expansion (2-fold greater than Dynabeads by day 14) despite a 413-fold lower antibody surface density [127]. The formation of an injectable T cell-responsive macroporous hydrogel is achieved through the cross-linking of DBCO-functionalized 8-arm PEG and azide-functionalized alginate via disulfide bonds, resulting in an average pore size of approximately 150 μm. The substantial size of the pores allows for the accommodation of T cells and the encapsulation of microspheres loaded with anti-CD3/CD28, thereby enabling the in situ activation of T cells. In a preventive model, this approach achieved 80% tumor progression-free survival in mice, demonstrating excellent antitumor efficacy [128]. In 2024, researchers have developed a 21 mm-diameter biomaterial scaffold based on a porous type I bovine collagen sponge [129]. By loading the chemokine CCL21 to recruit host T cells, the scaffold co-encapsulates a lentiviral vector encoding a tumor-specific chimeric antigen receptor (CAR) and a stimulatory anti-CD3/CD28 antibody, enabling in situ reprogramming and expansion of T cells [129]. In a mouse breast cancer model, 8 days after implantation, the proportion of CAR+ T cells at the tumor site reached 39.2 ± 19.5%, significantly higher than the 3.9 ± 1.1% in the spleen. Post-treatment tumor volume decreased by an average of 19.1%, and the median survival time of mice was extended to 43 days, outperforming traditional CAR-T therapy and the control group. A salient feature of this scaffold is its capacity to be stored at temperatures ranging from 2 to 8 °C following large-scale production, thereby satisfying clinical requirements [129]. In contrast to nanoscale particulate aAPCs, 3D platform systems provide a unique lymph node–mimicking mechanical microenvironment that supports sustained T cell stimulation, infiltration, and memory formation. They excel in ex vivo T cell expansion and local peri-tumoral immunotherapy rather than systemic administration. These specialized characteristics make platform systems irreplaceable in precise and localized immune modulation. Brief summary Based on the comparative analysis in Table 1, several key patterns emerge. For ex vivo T cell expansion, cell-based aAPCs (e.g., K562) and platform systems (e.g., 3D scaffolds) are the most favorable options, aligning with their high suitability scores in the table. Cell-based systems benefit from well-established genetic engineering protocols and phenotypic stability, enabling scalable and robust T cell expansion. Meanwhile, 3D platforms offer a physiologically relevant microenvironment that supports long-term culture and promotes memory T cell differentiation. For in vivo applications, natural and hybrid particle-based aAPCs stand out as the most suitable, owing to their excellent biocompatibility, low immunogenicity, and circulation stability. Hybrid particles leverage biomimetic membranes to evade immune clearance and achieve targeted delivery, while natural particles such as DC-derived exosomes demonstrate inherent biocompatibility. In contrast, synthetic particles (especially non-degradable variants) and unmodified cell-based aAPCs face greater challenges in vivo, including rapid clearance, potential off-target effects, and safety concerns. Regarding T cell activation efficiency, biosafety, and stability: synthetic particles offer the highest tunability and batch-to-batch consistency, while hybrid particles balance tunability with biomimetic properties, though their stability can vary with membrane sourcing and coating efficiency. Natural particle-based aAPCs excel in biocompatibility but suffer from moderate stability and low yields, limiting large-scale production. Platform systems provide unique advantages in delivering sustained mechanical and biochemical signals, though their clinical adoption is currently hindered by manufacturing complexity and, in many cases, the need for invasive implantation. Beyond these technical considerations, manufacturing cost and batch-to-batch reproducibility represent critical translational hurdles, particularly for natural and hybrid particle-based platforms that rely on biological membrane sourcing. These comparative insights underscore that the optimal aAPC design is highly context-dependent, guided by the specific therapeutic application (ex vivo vs. in vivo), target T cell population, and translational pathway. Researching application directions of aAPCs for tumor immunotherapy Based on the mechanisms by which APCs activate T cells and the principles of tumor immunotherapy, aAPCs exert their effects through four primary strategies: (1) promote T cell activation and proliferation; (2) overcome or modulate the immunosuppressive TME; (3) enable targeted delivery to tumor cells or tissues; and (4) combine with other therapeutic modalities (Fig. 3). Classifications of researching application directions of aAPCs for tumor immunotherapy (By Figdraw). The directions can be categorized into four main aspects: (1) Promoting T cell activation and proliferation by optimizing other therapeutic processes and stimulating T cell ex vivo and in vivo. (2) Employing aAPCs loaded with photosensitizers to induce immunogenic cell death (ICD) via tumor-localized ROS generation; or employing aAPCs loaded with catalysts to alleviate hypoxia by elevating local oxygen concentration. Both approaches ultimately enhance T cell infiltration and improve the TME. (3) Modulating aAPCs with magnetism, homing receptors on tumor cells membrane, and chemokines to facilitate targeted delivery. (4) Combining other therapies, such as anti-PD-1 immunotherapy, with tumor immunotherapy based on aAPCs to achieve multi-modality combination treatment. Application directions are summarized from studies reviewed in Section 3. Promote T cell activation and proliferation Most aAPCs designs are centered around one of these strategies, with “promoting T cell activation and proliferation” receiving the most extensive attention. This is typically achieved through three approaches: First, aAPCs can optimize other therapeutic processes. For instance, Metzloff et al. developed activating lipid nanoparticles (aLNPs) as aAPCs conjugated with CD3/CD28 antibody fragments, which effectively activated and transfected primary human T cells [60]. This significantly accelerated CAR-T cell production. While the approach centers on T cell activation, its ultimate aim is to enhance CAR-T cell therapy. This strategy highlights the value of aAPCs in streamlining and improving existing immunotherapy pipelines. Second, aAPCs are used for ex vivo activation and expansion of T cells. Lu et al. designed latex microspheres that were loaded with H-2Kb-Ig/pTRP2 dimers, anti-CD28 antibodies, 4-1BBL and CD83 [73]. These microspheres were found to efficiently and specifically expand pTRP2-specific CTLs in vitro via antigen-specific signalling and co-stimulatory molecules. Similary, Perica et al. designed paramagnetic iron-dextran nanoparticles loaded with pMHC and anti-CD28 antibodies [95]. These nanoparticles exhibit excellent biocompatibility while utilizing their magnetic separation capability to achieve enrichment of antigen-specific components. Additionally, the dual signaling effectively promotes cell proliferation. The key technical difference between these two platforms lies in their material and functional orientation: latex microspheres achieve high-efficacy expansion, while magnetic nanosystems enable precise enrichment of antigen-specific T cells. Third, aAPCs can stimulate T cells in vivo. Majedi et al. constructed aAPCs with covalently bound anti-CD3 and anti-CD28 antibodies to deliver the first and second signals of T cell activation [125]. The mechanical stiffness of the scaffold was tuned to approximately 50 kPa to mimic lymph node elasticity, enhancing T cell mechanotransduction. Besides, IL-2 release supported T cell clonal expansion and survival. The chemokine CCL21 recruited effector T cells to the site, enabling localized activation and precise antitumor responses. This design mimics the natural lymph node microenvironment, achieving sustained and site-specific T cell activation, which is superior to simple transient stimulation. Collectively, the above studies reveal three important principles for aAPC design: (Ⅰ) multi-signal coordination is essential for effective T cell activation; (Ⅱ) material and structural properties directly determine activation efficiency and application scenarios; (Ⅲ) ex vivo and in vivo applications require distinct platform design strategies. These advances lay a critical foundation for aAPC-based personalized T cell immunotherapy. Notably, although IL-2 is widely used in aAPC systems to support T cell growth, it can preferentially expand regulatory T cells (Tregs) due to their high-affinity CD25 receptor, potentially weakening antitumor immunity. As an alternative, IL-15 and IL-21 have emerged as favorable cytokine candidates, as they preferentially promote effector T cell and memory T cell proliferation without excessive Treg activation. Overcome or modulate the immunosuppressive TME Recently, there has been increasing interest in research about integrating several therapeutic mechanisms. Among them, it is significant to emphasize that modulation of the immunosuppressive TME is frequently mentioned to combine with other strategies of T cell activation. For example, Sun et al. engineered membrane-coated nanoparticles called saDC@Fs-NPs, which were designed by wrapping photosensitizer-loaded particles with saDC membranes displaying pMHC-I, CD86, and anti-LAG3 antibodies [117]. This system both activated T cells and blocked the MHC-II/LAG3 exhaustion pathway. Upon light stimulation, the photosensitizer induced ROS-mediated immunogenic cell death (ICD), releasing damage-associated molecular patterns (DAMPs) that activated DCs and facilitated T cell infiltration—effectively integrating T cell activation and TME modulation. Similarly, Lu et al. constructed aAPCs called CD-MnOx@CM by coating MnOx-NH2 nanoparticles with B16-F10 melanoma cell membranes conjugated with anti-CD3/CD28 antibodies [118]. These aAPCs achieved tumor targeting via homotypic membrane recognition, activated T cells, and improved the TME by decomposing hydrogen peroxide and mitigating hypoxia. However, their reliance on tumor-specific membrane markers may limit efficacy in heterogeneous tumors, and the long-term accumulation of Mn2+/Mn3+ poses safety concerns requiring further study. Further, based on the research mentioned above, many biomimetic aAPCs designs integrate multiple therapeutic pathways, particularly the combination of T cell activation and TME modulation. By embedding co-stimulatory molecules or pMHCs in the biomimetic membrane, they facilitate T cell activation and proliferation. Simultaneously, the nanoparticle core actively alters the TME through its physicochemical properties or controlled release of agents. This integrated strategy significantly enhances antitumor immunity by reversing immune suppression and boosting T cell function. In addition, activation of the STING pathway has emerged as a promising strategy to remodel the immunosuppressive TME. STING agonists trigger type I interferon responses, promote dendritic cell maturation, and facilitate cross-priming of T cells, thereby converting immunologically “cold” tumors into “hot” tumors [56,113]. Although the direct incorporation of STING agonists into aAPC platforms is still evolving, this approach offers a complementary mechanism to enhance T cell activation and infiltration, and represents a valuable direction for future aAPC-based combination immunotherapy. Self-assembling PD-L1-targeting peptides can enhance γδ T cell activity and alleviate immunosuppression in the TME [130]. Organelle-targeted polymers trigger immunogenic pyroptosis, effectively turning cold tumors into hot tumors [131]. Enable targeted delivery to tumor cells or tissues Targeted therapy is achieved through physical properties (e.g., magnetism) or biological cues (e.g., tumor homing and chemokine guidance). Sun et al. leveraged the natural expression of homing receptors (e.g., L-selectin) on lymphocytes to develop LC-aAPCs [45]. Upon intravenous administration, these aAPCs migrated to secondary lymphoid organs, such as the spleen and lymph nodes. In these organs, they enriched and enhanced spatially specific T cell activation. This effect was confirmed by in vivo imaging and flow cytometry. Combine with other therapeutic modalities In addition, aAPCs can be used in combination with other therapies when employed in tumor immunotherapy. Kosmides et al. combined aAPCs with PD-1 monoclonal antibodies to block the PD-1/PD-L1 immune checkpoint pathway within the TME [19]. This approach significantly delayed melanoma progression and prolonged survival in mouse models. Actually, numerous other studies have similarly explored the synergistic effects of this approach [45,69,120]. Additionally, Butler et al. proposed a combination of anti-CTLA-4 monoclonal antibody ipilimumab and aAPCs [46]. In essence, however, the mechanism of these studies involves the activation of antigen-specific T cells by aAPCs, accompanied by the release of immunosuppression through the administration of immune checkpoint inhibitors. This combination is intended to achieve a synergistic antitumor effect by combining activation and potentiation. Meanwhile, another research demonstrated a synergistic promotion of efficient enrichment, expansion, and functional maintenance of γδ T cells through the combination of aAPCs with zoledronic acid (ZOL) and IL-2 [29]. Bioactive poly(amino acid) nanomaterials enable synergistic integration of ICD, cancer vaccination, and immune checkpoint blockade [132]. Chemotherapeutic agents such as doxorubicin can induce ICD of tumor cells, leading to the release of DAMPs and increased tumor antigen presentation. When combined with aAPCs, chemotherapy enhances the availability of tumor antigens, while aAPCs sustain the activation and proliferation of tumor-specific T cells. This synergistic strategy converts localized chemotherapy into systemic antitumor immunity. A size-transformable aAPC nanovesicle system has been developed in combination with doxorubicin, showing improved antitumor efficacy in solid tumor models while reducing systemic toxicity [104]. Photodynamic therapy generates reactive oxygen species (ROS) to trigger ICD and remodel the TME. aAPCs integrated with photosensitizers can simultaneously achieve T cell activation and photodynamic immunotherapy. For example, membrane-coated aAPCs loaded with AIE-based photosensitizers enable near-infrared-controlled immune activation, enhancing T cell infiltration and reversing immunosuppression. This spatiotemporally controllable combination is minimally invasive and suitable for localized solid tumors [117]. Radiotherapy induces ICD, which can trigger anti-tumor immune responses and reshape the TME [133]. When combined with aAPCs, radiation-induced antigen release could be coupled with robust T cell priming by aAPCs, thereby amplifying systemic antitumor responses. Scaffold-based aAPCs have been shown to support efficient ex vivo expansion of functional T cells [123], and may be further combined with local radiotherapy to enhance T cell retention and activity, representing a promising strategy to improve outcomes in radio-resistant tumors. Targeted therapies, such as tyrosine kinase inhibitors (TKIs), can selectively inhibit oncogenic signaling and reduce immunosuppression within the TME [134]. Combining such targeted agents with aAPCs may further enhance tumor clearance and boost the persistence and function of aAPC-activated T cells, leading to more durable antitumor immunity. Current clinical applications of aAPCs and translational bottlenecks The clinical application of aAPCs remains limited, with most investigations still restricted to preclinical animal studies. In 2011, an open-label, single-arm Phase I dose-escalation trial led by Butler and colleagues evaluated the administration of engineered aAPCs in nine patients with advanced melanoma [20]. Following injection, various immunological parameters, including the quantity and phenotype of peripheral blood CTLs, are periodically monitored. The frequency of MART1-specific CD8+ T cells in the peripheral blood of all patients tested was significantly higher, with a median increase of 6.1-fold in the low-dose group and 9.8-fold in the high-dose group, yielding encouraging preliminary results. However, the small sample size (n = 9), lack of a control arm, and the exploratory nature of this feasibility-focused Phase I design preclude definitive efficacy conclusions. Beyond this pioneering clinical experience, the potential clinical value of aAPCs can be further appreciated when compared with currently established immunotherapies. Unlike CAR-T therapy, which relies on viral vector-mediated gene transduction and carries risks of insertional mutagenesis and prolonged manufacturing timelines, aAPCs—especially synthetic and hybrid platforms—offer the prospect of “off-the-shelf” availability, shorter production cycles, and reduced safety concerns. Compared to DC vaccines, aAPCs exhibit superior batch-to-batch consistency and modular tunability, circumventing the limitations of autologous DC isolation and variability. Relative to immune checkpoint inhibitors (ICIs), aAPCs provide direct T cell co-stimulatory signals, potentially initiating de novo anti-tumor immune responses in immunologically “cold” tumors that respond poorly to ICIs alone. Collectively, these attributes position aAPCs as a promising complementary modality in the clinical immunotherapy landscape. However, the small sample size and the requirement for combination therapy in most patients highlight the study's limitations. The aAPCs used in this trial are derived from a platform previously developed by the authors and described in a 2007 Clinical Cancer Research publication [24]. This platform employs the K562 cell line, genetically modified to overexpress HLA-A2, CD80, and CD83 on the cell surface. Such a design necessitates patient-specific preparation, significantly increasing production costs and limiting scalability. More than a decade after this first-in-human trial, the clinical translation of aAPCs remains in its early stages, with large-scale clinical trial data still scarce. The limited progress reflects several interconnected challenges that span patient heterogeneity, safety uncertainties, and manufacturing complexity—issues that are elaborated in detail in the following section. Challenges and imperfections Although aAPCs exhibit broad potential in tumor immunotherapy, several critical challenges impede their large-scale clinical application. First, therapeutic efficacy remains unpredictable due to patient-to-patient variability. As highlighted by Butler et al. [20], some patients still require combination therapies to achieve disease control, indicating that identical aAPCs may elicit divergent outcomes across individuals. Such variability is likely attributable to differences in endogenous antigen profiles, complicating the optimization of dosing, administration frequency, and treatment protocols. On the other hand, it should be noted that due to the low proportion of central memory T cells (Tcm) and the rapid depletion of effector T cells, the efficacy of tumor immunotherapy based on aAPCs is not enduring. Second, the long-term risks and adverse effects associated with aAPC-based therapies remain poorly characterized. Most studies to date are restricted to preclinical animal models, where biosafety profiles are generally favorable. Nevertheless, the potential risks associated with prolonged administration require further investigation. For example, DNA microflowers (DM) engineered by Le et al. [88] effectively inhibit both primary and metastatic tumors, induce systemic immune responses, and prolong murine survival. Fluorescence imaging reveals predominant accumulation in tumor tissues, with minimal and declining signals in vital organs such as the heart, liver, spleen, lungs and kidneys, thereby supporting their biocompatibility. However, these findings are limited to animal models so the potential risks remain insufficiently explored. For example, there is the risk of chronic toxicity from DNA degradation products or polydopamine coatings. Additionally, there is also the risk of chronic inflammation, caused by long-term interactions between interleukin-2-loaded antibody-modified DNA microflowers (DM-AI) and the immune system. Potential off-target effects, such as unintended activation of non-specific T cells or accumulation in non-target organs, also warrant systematic evaluation. Third, the absence of standardized and scalable production protocols for aAPCs represents a substantial barrier to clinical translation. Despite the rapid emergence of diverse aAPC constructs, no industry consensus or expert guidelines currently exist to regulate their manufacturing and quality control. For instance, Dong et al. [90] report that the instability of microemulsion droplets during the fabrication of polydopamine microbubbles results in poor particle size uniformity, a limitation recognized as a key direction for future improvement. Future prospects Advancing aAPC technology requires a deeper understanding of how size, morphology, surface topography, and material composition influence immunostimulatory efficacy. Research has demonstrated that particle size critically impacts T cell activation kinetics. For example, SPIONs exceeding 300 nm facilitate effective engagement with multiple TCR nanoclusters, leading to robust proliferation, whereas smaller counterparts (∼50 nm) at equivalent ligand doses elicit significantly weaker responses [85]. Particle shape also plays a significant role. Non-spherical geometries (e.g., ellipsoids) often outperform spheres in activation efficiency. This may be due to an enhanced interfacial contact area, altered membrane dynamics that influence ligand clustering, or differential in vivo trafficking and circulation times [[61], [62], [63]]. Surface characteristics such as ligand density, spatial arrangement (achieved with DNA origami, which enables nanoscale precision [87]), and membrane fluidity (exploited in liposomal and biomimetic designs [57,59,114]), directly modulate TCR triggering strength, synapse stability, and signal integration. The choice of material not only affects biocompatibility and biodegradability, but also influences drug loading/release profiles, mechanical properties (e.g., scaffold stiffness that mimics lymph nodes [125]), and potential off-target effects. The future design of aAPCs will require a systematic, AI-assisted exploration of these multidimensional parameter spaces. This exploration will identify configurations that maximize T cell activation, expansion, functional differentiation, tumor homing, and persistence, while minimizing exhaustion and unintended immunomodulation. Chiral material engineering provides a novel strategy to enhance antigen presentation and T cell activation [135]. Intrinsically immunomodulatory biomaterials that activate the STING pathway can further boost the efficacy of aAPC-based immunotherapy [136]. Firstly, the role of AI in the design of aAPCs is anticipated to become increasingly significant. For example, as mentioned earlier, in a study by Ben-Akiva et al. [62], the impact of aAPC morphology on T cell activation was investigated. The researchers elongated spherical aAPCs into prolate and oblate ellipsoids and discovered that both ellipsoid shapes outperformed the spherical form in activation efficiency. What's more, the oblate ellipsoid exhibited superior external binding and uptake, whereas the prolate ellipsoid showed prolonged circulation time in vivo. Given this, could AI be utilized to further refine morphological design to maximize efficiency? Indeed, there have already been pioneering studies demonstrating the power of AI in optimizing nanocarrier design for biomedical applications. A prime example is the use of deep learning, specifically a directed message-passing neural network (D-MPNN) termed LION (Lipid Optimization using Neural Networks), to revolutionize the discovery of ionizable lipids for lipid nanoparticles (LNPs). This approach leveraged a substantial dataset of over 9000 LNP activity measurements to train the model, enabling the in silico screening of a staggering 1.6 million potential lipid structures [137]. Notably, the AI model effectively identified novel lipid structures, including FO-32 and FO-35, which exhibited notable performance divergent from conventional design heuristics. For instance, FO-35 exhibited a distinctive perillyl alcohol tail, a feature not previously observed in other lipids. The investigation revealed that the lipids designed with AI exhibited high potency in vitro and in mouse models. For instance, they matched or exceeded state-of-the-art benchmarks, such as SM-102, for intramuscular mRNA delivery. Additionally, they demonstrated superior transfection in mouse lung and nasal mucosa via optimized formulations, including F3, a LNP formulation optimized by design of experiments. Furthermore, these lipids achieved remarkably efficient and widespread transfection throughout the ferret lung epithelium upon nebulized mRNA delivery. For example, the FO-32 lipid achieved an average of approximately 60% GFP+ epithelial cells in the transfected bronchi. In comparison, the leading control lipid, IR-117-17, achieved an average of approximately 1.4% GFP+ epithelial cells. Furthermore, key physical-chemical properties relevant to delivery efficiency and stability were characterized for these AI-optimized LNPs, including pKa values (e.g., FO-32 pKa ∼6.36, within the optimal range for endosomal escape) and resilience to freeze-thaw cycles with minimal changes in particle diameter. These findings underscore the capability of AI to navigate vast chemical and formulation spaces, identifying non-intuitive yet highly effective designs. This strongly suggests that similar AI-driven strategies could be transformative for aAPC development, potentially identifying the optimal physical parameters, including aspect ratio, surface topography, ligand density, and mechanical properties. It has the potential to optimize critical interactions with immune cells (e.g., T cell binding, synapse stability, and activation kinetics), thereby substantially augmenting the comprehensive therapeutic efficiency of aAPCs. Secondly, the implementation of more diversified combination therapy holds considerable potential for further research. For now, the majority of research on combination therapy continues to prioritize the combination of aAPCs and immune checkpoint inhibitor therapy. However, it is important to note that there are numerous other existing therapeutic options for tumors, including radiation therapy, surgical therapy, and gene therapy. A future research objective will be to determine the most effective methods of combining aAPCs with these therapies. For instance, whether aAPC therapy can be combined with surgical treatment is one of valuable directions. With the combination of aAPCs and surgery, larger tumors could be treated with aAPC initially to reduce their size before undergoing surgery. Besides, a further inquiry pertains to the potential utilization of aAPC in the context of tumors that are difficult to completely remove through surgery, including ovarian cancer. This will require the exploration of more diversified diagnostic and therapeutic pathways, with the aim of improving the treatment of tumors and prolonging patient survival. Third, the diversification of aAPC designs is identified as a pivotal area of focus. Stimulus-responsive aAPCs represent an exciting field for future development. Yang et al. designed a size-variable aAPC that incorporates redox-sensitive materials on its surface, which specifically interact with the T cell microenvironment to trigger activation [104]. This design suggests the possibility of developing aAPCs that respond to various external stimuli, including light, pH, temperature, and ultrasound [55,56] to induce morphological changes, drug release, or other effects. By controlling these variables, aAPCs functionality could be finely tuned, enabling more precise and personalized therapeutic approaches. Besides, the development of multifunctional aAPCs also represents a significant research focus. Beyond facilitating tumor immunotherapy, these artificial cells are investigated for their potential to perform additional roles, such as enhancing tissue regeneration. The integration of multiple therapeutic capabilities within aAPCs is anticipated to lower the overall cost of disease management and mitigate patients’ financial burden. Another promising area of research is developing aAPCs that can reverse T cell exhaustion. Studies have shown that terminally exhausted T cells may retain transcriptional capacity under conditions of reduced hypoxia and enhanced co-stimulatory signals. These findings suggest that reversing T cell exhaustion may enable more effective immunotherapy against tumors [138]. In fact, research teams have designed drugs to reverse T cell exhaustion. For instance, a fusion protein consisting of an anti-TIM3 antibody and an IL-2 precursor can specifically target exhausted CD8+T cells in the TME [139]. This enables the precise delivery of IL-2. Through the “bystander effect”, the fusion protein activates the TIM3- TIL population, including PD-1+ TIM3- stem cell-like and early effector T cells. This raises the question of whether aAPCs could be combined with mechanisms that reverse T cell metabolism to enhance the efficacy of immunotherapy against tumors. Additionally, the development of more individualized aAPC designs is a critical area of research. As demonstrated in the clinical study conducted by Butler et al. [20], individuals exhibit divergent responses when confronted with the same aAPC, and the efficacy of aAPC varies significantly among different individuals. A substantial amount of studies have demonstrated that when there are differences in ethnicity, population, cancer type, region, and sex, tumor cells exhibit significant variability across multiple dimensions [[140], [141], [142], [143], [144]]. At present, a growing number of antigen designs are emphasizing individual specificity to a greater extent than ever before [[145], [146], [147]]. Consequently, the development of aAPCs tailored to individual patients represents a promising avenue for further research. Fourth, the incorporation of novel technologies is likely to accelerate advancements in aAPCs research. For example, El Hebieshy et al. introduced “ubi-tagging,” a multivalent coupling method that efficiently links antibodies, antibody fragments, nanobodies, peptides, or small molecules to antibodies and nanobodies [148]. This technology has the potential to enhance the coupling efficiency of aAPCs while maintaining the stability and functionality of the antibodies, thus advancing their therapeutic potential. Finally, in order to evade attack by the immune system, tumor cells undergo rapid and variable mutations. In the future, it may be possible to design an “autonomously evolving” aAPC that can automatically sense and track changes in tumor cells and make targeted changes itself, thereby more efficiently presenting the corresponding antigen signals. Conclusions Artificial antigen-presenting cells (aAPCs) have emerged as a promising technological advancement in tumor immunotherapy, gradually overcoming the limitations of traditional natural APCs by leveraging their multifunctional capabilities in T cell activation, TME modulation, and precise tumor targeting therapy. In recent years, various types of aAPCs, engineered using cell-based strategies, materials, biomimetic approaches, and aAPCs platform systems, have demonstrated substantial immune-regulatory effects in both in vitro and in vivo models, thereby contributing to the advancement of personalized and precision medicine strategies. Despite significant progress in the functional design and application of aAPCs, their clinical translation remains hindered by patient-specific biological variability, potential biosafety concerns during long-term use, and the lack of standardized production and quality control systems. In the future, it is essential to leverage AI-assisted design, diversified combination therapy, stimulus-responsive materials, multifunctional aAPCs and the application of innovative technologies to further enhance the functional precision, controllability, and safety of aAPCs. Moreover, the continued optimization and widespread application of aAPCs depend on the integration of interdisciplinary fields, such as biomaterials science, immunology, oncology, synthetic biology, and data science. Strengthening interdisciplinary collaboration and incorporating cutting-edge technologies and novel ideas will be essential to addressing existing challenges, expanding the scope of aAPCs in tumor immunotherapy, and advancing their clinical standardization and large-scale application. Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Funding This work was supported by the National Key Research and Development Program of China (2024YFA0917700), the Shanghai Municipal Commission of Science and Technology (25TS1412200), the National Natural Science Foundation of China (Grant No. 82272888), Shanghai Hospital Development Center Foundation (Grant No. SHDC12022106, 2022SKLY-12), Tongji University (2026-0555-YB-05), Shanghai First Maternity and Infant Hospital (Grant No. 2023B01、2023A01、2024A05、2024B20) and Health Commission of Shanghai Pudong District (Grant No. PWZxq2022-03、PW2023D-14). CRediT authorship contribution statement Haojie Cai: Data curation, Formal analysis, Methodology, Writing – original draft. Yaqian Zhao: Investigation, Project administration, Validation. Yifei Cai: Data curation, Formal analysis, Investigation. Xiaomei Luo: Data curation, Formal analysis, Methodology. Jianxiao Li: Data curation, Formal analysis, Project administration. Yongsong Wu: Data curation, Formal analysis, Investigation. Jiani Yang: Methodology, Validation. Chao Wang: Investigation, Project administration. Yue Zhang: Investigation, Project administration. Hui Ding: Investigation, Validation. Shanshan Cheng: Conceptualization, Data curation, Writing – review & editing. Yu Wang: Conceptualization, Data curation, Funding acquisition, Writing – review & editing. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements We sincerely thank the free drawing support provided by the Figdraw platform (www.figdraw.com). Figs. 1 and 3 in the paper were drawn by Figdraw. Contributor Information Haojie Cai, Email: chjlcyx@163.com. Yaqian Zhao, Email: yxszhaoyaqian@163.com. Yifei Cai, Email: cyf991134@163.com. Xiaomei Luo, Email: luoxiaomei26@163.com. Jianxiao Li, Email: lijianxiao@tongji.edu.cn. Yongsong Wu, Email: wuyongsong@51mch.com. Jiani Yang, Email: sjtuyangjiani@163.com. Chao Wang, Email: 13761183836@163.com. Yue Zhang, Email: yzhang2013@126.com. Hui Ding, Email: dinghui_doctor@163.com. Shanshan Cheng, Email: doctorcheng@alumni.sjtu.edu.cn. Yu Wang, Email: renjiwangyu@126.com. List of abbreviations antigen-presenting cells artificial APCs tumor microenvironment artificial intelligence chimeric antigen receptor T cell chimeric antigen receptor immune checkpoint blockade programmed cell death protein 1 programmed death-ligand 1 adoptive cell therapy tumor-infiltrating lymphocyte dendritic cells major histocompatibility complex peptide-MHC T cell receptor interleukin-12 interleukin-18 cytotoxic T lymphocytes helper T cells lymphocyte function-associated antigen-1 Fas ligand interferon-gamma interferon-alpha natural killer magnetic nanoclusters lymphocyte-based homologous targeting artificial antigen-presenting cells Phosphate Buffered Saline Antigen-Presenting Cytochalasin-B-Induced Membrane Vesicles peripheral blood mononuclear cells lipid nanoparticles activating LNPs poly(lactic-co-glycolic acid) poly(beta-amino-ester) poly(N-isopropylacrylamide) blood–brain barrier superparamagnetic iron oxide nanoparticles two-dimensional three-dimensional rolling circle amplification super artificial DCs artificial T cell-stimulating matrices hyaluronic acid polyethylene glycol diacrylate artificial lymph nodes magnetic resonance imaging polycaprolactone microfibers regulatory T cells immunogenic cell death damage-associated molecular patterns zoledronic acid reactive oxygen species tyrosine kinase inhibitors immune checkpoint inhibitors DNA microflowers interleukin-2-loaded antibody-modified DNA microflowers directed message-passing neural network Lipid Optimization using Neural Networks Data availability No data was used for the research described in the article. References Associated Data Data Availability Statement No data was used for the research described in the article.