Advancing CAR-based cell therapies for solid tumours: challenges, therapeutic strategies, and perspectives

Abstract

Chimeric antigen receptor-cell therapies have demonstrated remarkable success in haematological malignancies but face significant hurdles in solid tumours. The hostile tumour microenvironment, antigen heterogeneity, limited tumour infiltration, and CAR-cell exhaustion contribute to reduced efficacy. Additionally, toxicity, off-target effects, and manufacturing challenges limit widespread clinical adoption. Overcoming these barriers requires a multifaceted approach that enhances CAR-cell persistence, trafficking, and tumour-specific targeting. Recent advancements in alternative cellular therapies, such as CAR-natural killer cells, CAR-macrophages, gamma delta CAR-T cells, and CAR-natural killer T cells, provide promising avenues for improving efficacy. These strategies leverage distinct immune cell properties to enhance tumour recognition and persistence. Furthermore, combination therapies, including chemotherapy, radiotherapy, antibodies, small molecule inhibitors, cancer vaccines, oncolytic viruses, and multi-CAR cell combination therapy, offer synergistic potential by modulating the TME and improving CAR-cell functionality. This review explores the challenges of CAR-based cellular therapies in solid tumours and highlights emerging strategies to overcome therapeutic limitations. By integrating novel cellular platforms and combination approaches, we seek to provide insights into optimising CAR-cell therapies for durable responses in solid malignancies.

Keywords: CAR-Macrophages; CAR-NK cells; CAR-T cells; Challenges; Combination therapy; Optimisation Strategies; Solid tumours.

Fig. 1

The basic structure and evolution of CARs across generations. A Schematic representation of the canonical CAR architecture, consisting of an extracellular single-chain variable fragment (scFv) derived from tumour antigen-specific antibodies for antigen recognition, a hinge or spacer region for conformational flexibility, a transmembrane domain (commonly derived from CD4, CD8α, or CD28) for membrane anchoring, and an intracellular signalling module. The intracellular domain includes the CD3ζ chain, containing immunoreceptor tyrosine-based activation motifs (ITAMs) for T-cell activation, often in combination with co-stimulatory domains [5, 6]. B Outlines the evolution of CAR generations. First-generation CARs consist solely of the CD3ζ intracellular signalling domain, which initiates T-cell activation but lacks co-stimulatory signals necessary for robust in vivo function [6]. Second-generation CARs introduce a single co-stimulatory domain—most commonly CD28 or 4-1BB—enhancing T-cell expansion, persistence, and anti-tumour efficacy. CD28-based CARs drive rapid effector responses, while 4-1BB confers superior metabolic fitness and durability [7, 8]. Third-generation CARs incorporate two co-stimulatory domains (e.g., CD28 + 4-1BB), offering synergistic enhancement of cytokine production and cytotoxicity [9, 10]. Fourth-generation CARs, or TRUCKs (T cells Redirected for Universal Cytokine-mediated Killing), include inducible cytokine-expression modules (e.g., IL-12, IL-15, IL-18) that modulate the immunosuppressive TME and improve local immune activation [11]. Fifth-generation CARs build on second-generation frameworks by integrating a truncated IL-2Rβ chain with a STAT3/5-binding motif, enabling direct activation of the JAK–STAT signalling cascade to enhance T-cell proliferation, memory differentiation, and long-term anti-tumour responses [12]

Fig. 2

This figure provides a comprehensive overview of the mechanisms that enable tumours to evade immune surveillance and resist immune-mediated destruction, forming a critical framework for understanding immune resistance in cancer. A The TME plays a central role in promoting immunosuppression through various interconnected processes. Tumours create an immunosuppressive niche by promoting the expansion of Tregs, MDSCs, and CAFs, which secrete inhibitory cytokines like TGF-β and IL-10 to suppress cytotoxic immune responses. B Immune checkpoint pathways, including PD-L1, CTLA-4, TIM-3, and LAG-3, further contribute to immune evasion by inducing T-cell exhaustion and impairing CTL responses. C In addition, tumours escape immune recognition by downregulating major MHC-I molecules and NK-cell-activating ligands (e.g., NKG2DL), reducing antigen presentation and cytolytic activity. D Beyond direct immune modulation, tumour cells and stromal components secrete immunosuppressive and pro-tumourigenic factors such as IL-10, TGF-β, PGE2, and VEGF, which inhibit T-cell and NK-cell function while promoting angiogenesis and disease progression. E Resistance to apoptosis is another critical factor, where genetic and epigenetic alterations, including p53 loss and overexpression of survival proteins like BCL-2 and survivin, disrupt cell death pathways, making tumour cells resistant to immune-mediated killing. F Furthermore, metabolic reprogramming within the TME, such as the Warburg effect and increased lactate production, leads to an immunosuppressive metabolic landscape that further inhibits T-cell and NK-cell function. G Lastly, tumour plasticity, driven by EMT and cancer stem cell (CSC) survival, enhances immune evasion, therapy resistance, and metastatic potential. Together, these mechanisms create a formidable barrier to effective immunotherapy, highlighting the need for novel strategies to counteract immune resistance in cancer treatment

Fig. 3

This illustration presents advanced car designs, categorised into multi-target cars and modular car constructs with optimised function, specificity, and safety. A The Multi-Target CARs are designed to enhance antigen recognition and tumour targeting by incorporating multiple antigen-binding domains. These include Bicistronic CARs, which express two independent CAR constructs within a single T cell to target distinct antigens; Bispecific-Tandem CARs, which contain a single receptor with dual antigen-binding domains; Trivalent CARs, which incorporate three antigen-binding sites to increase specificity and reduce tumour escape; and Loop CARs, which feature a looped scFv structure to improve antigen-binding flexibility. B The Modular CAR Constructs integrate additional regulatory mechanisms to refine CAR-T cell activation, specificity, and safety. iCasp9 Small-Molecule-Controlled CARs include an inducible caspase-9 suicide switch that allows for the controlled depletion of CAR-T cells in case of severe toxicity. Reversed (Rev) CARs require a bispecific bridging molecule for activation, preventing unwanted tonic signalling. NOT Logic Gate CARs (iCARs) integrate inhibitory receptors that suppress activation upon recognising an off-target antigen, reducing damage to healthy tissues. IF-Better Logic Gate CARs introduce an activation threshold that ensures T-cell activation only in response to high antigen expression, preventing reactivity to normal tissues with low antigen levels. Avidity (Avid) CARs require bivalent antigen binding for full activation, preventing responses to cells with low antigen expression. AND-Gate CARs incorporate two separate receptors, where both antigens must be engaged to trigger full T-cell activation, ensuring strict tumour selectivity. Synthetic Notch (synNotch) CARs enable sequential antigen recognition, where binding to a first antigen induces the expression of a second CAR, allowing for delayed and more precise T-cell activation. Lastly, SUPRA-CARs (Split, Universal, and Programmable CARs) utilise a universal receptor system with modular adapter molecules, allowing for real-time reprogramming of antigen specificity without requiring genetic modification of T cells. These next-generation CAR constructs provide enhanced precision, flexibility, and safety in CAR-T cell therapy, making them promising candidates for improved treatment of both haematological malignancies and solid tumours

Fig. 4

This figure illustrates potential strategies to overcome the challenges of CAR-T cells in solid tumours. Several barriers hinder the efficacy of CAR-T cell therapy in solid malignancies, necessitating novel strategies for optimisation. A Immunosuppressive TME: Engineering CAR-T cells with HIF-1α knockout or TAGLN2 overexpression enhances their function in hypoxic conditions. Macrophage reprogramming via CSF1R inhibition and MDSC modulation with IDO/arginase inhibitors restores immune-activating conditions. Treg depletion using anti-CTLA-4 strategies further improves immune response. B Poor Tumour Infiltration: Strategies include chemokine receptor expression matching tumour-derived chemokines, ECM degradation via heparanase-expressing CAR-Ts, and combination therapies with chemotherapy and radiotherapy to improve homing and penetration. C Antigen Heterogeneity: Multi-target CAR designs such as bicistronic, bispecific-tandem, trivalent, and loop CARs address tumour antigenic variability and escape mechanisms (Refer to Fig. 3). D CAR-T Exhaustion: Enhancing persistence through immune checkpoint receptor knockout (LAG-3, PD-1, TIM-3), cytokine modulation (IL-15/IL-21), and alternative cell sources with longer persistence abilities like γδ T and NKT cells. E Toxicity & Off-Target Effects: Modular CAR constructs optimise specificity and safety, including small-controlled, reversible, synthetic Notch, avidity-based, and logic-gated CARs. F Manufacturing Challenges: Overcoming limitations using alternative cell sources (iPSC-derived, allogeneic “off-the-shelf” CAR-Ts, and MAIT cells) and in vivo CAR-T generation approaches. These advancements collectively improve CAR-T cell therapy outcomes in solid tumours

Fig. 5

This figure illustrates the major challenges of CAR-NK cell therapy in solid tumours and the strategies designed to address these obstacles. A The first challenge is the lack of persistence, where CAR-NK cells exhibit limited durability due to the immunosuppressive TME. Factors such as metabolic stress (hypoxia, low pH, nutrient depletion), ROS, and inhibitory cytokines (TGF-β, PGE2, IL-10, IL-16) contribute to CAR-NK cell exhaustion, while suppressive immune cells, including TAMs, MDSCs, and Tregs, further inhibit CAR-NK activity. Strategies to overcome this limitation include sourcing NK cells from diverse origins such as peripheral blood, umbilical cord blood, or stem cells, cytokine preconditioning (IL-2, IL-12, IL-15, IL-18) to enhance CAR-NK memory-like function, utilising biodegradable scaffolds and hydrogels for in vivo support, and engineering CAR-NK cells for survival and resistance by targeting adenosine and TGF-β signalling or overexpressing pro-survival factors like Bcl-2, Bcl-XL, and Mcl-1. The second challenge is antigen heterogeneity and off-target effects. To mitigate these effects, precision CAR-NK engineering can target specific tumour markers, and logic-gated or suicide gene systems, such as inducible caspase-9 (iCasp9), can be employed to prevent fratricide and minimise off-target toxicity. The third challenge, manufacturing limitations, stems from the inefficiency and cost of genetically modifying NK cells using viral (e.g., lentivirus) or non-viral systems. Strategies to optimise manufacturing include refining expansion protocols with IL-15 and cytokine cocktails to improve CAR-NK proliferation, as well as enhancing gene delivery methods through electroporation, lipid nanoparticles, multifunctional nanoparticles, and cell-penetrating peptides to improve transduction efficiency

Fig. 6

This figure highlights six key challenges in chimeric antigen receptor macrophage (CAR-M) therapy and corresponding strategies to enhance anti-tumour efficacy. A Limited Tumour Infiltration: CAR-Ms struggle to penetrate tumours due to the ECM. Engineering chemokine receptors (CCR2, CXCR3) and ECM-degrading enzymes (heparanase, hyaluronidase) improve infiltration. B Limited Persistence: CAR-M longevity is enhanced through PD-1 knockout, metabolic reprogramming (HIF-1α knockout, lactate transporter inhibition), and iPSC-derived macrophages with improved renewal pathways (STAT3/c-Myc). C Phenotype Plasticity: Preventing M2 polarisation via IL-10R/TGF-βR knockout and reinforcing M1 activation with pro-inflammatory cytokines (GM-CSF, IFN-γ, IL-12) enhances CAR-M functionality. Hydrogel/nanoparticle delivery protects CAR-Ms from immunosuppressive TME signals. D Gene Transfer Challenges: Efficient gene delivery is achieved via electroporation-based mRNA transfection, lipid nanoparticles, optimised viral vectors, and biomaterial-assisted stable delivery. E Toxicity & Off-Target Effects: Safety is improved with tunable CAR systems, including drug-inducible and suicide switch-controlled CARs (e.g., iCasp9). Local CAR-T administration further reduces systemic toxicity. F Limited Antigen Presentation: Enhancing antigen presentation involves generating macrophage-dendritic cell hybrids (upregulating MHC-II, CD80, CD86) and CAR-M-derived exosomes to prime T cells. These innovations aim to optimise CAR-M therapy for improved tumour targeting, persistence, and safety

Fig. 7

The multi-CAR-cell combination strategy harnesses the complementary functions of CAR-T cells, CAR-NK cells, and CAR-Ms to overcome the complex barriers presented by solid tumours. CAR-T cells, genetically modified to target TAAs, are central to the cytotoxic response. Once activated, they proliferate, secrete effector molecules, and directly lyse tumour cells. However, in the context of solid tumours, CAR-T cells face several limitations, including poor infiltration into the TME, antigen heterogeneity, and functional exhaustion driven by chronic stimulation and immunosuppressive cues. To enhance their activity, CAR-NK cells are co-administered. These cells exhibit rapid, antigen-independent killing via NKG2D–NKG2DL interactions and CD16-mediated ADCC, and importantly, they secrete cytokines such as IFN-γ and TNF-α, which stimulate both CAR-T cells and innate immunity. Their presence helps remodel the TME and improves CAR-T cell recruitment and function. CAR-Ms further support this cooperative network. By infiltrating the tumour mass, phagocytosing cancer cells, and shifting towards an M1-like phenotype, they promote inflammation, antigen presentation, and T-cell recruitment. Through secretion of chemokines and cytokines, including IL-12 and CCL3/CCL5, CAR-Ms enhance T cell infiltration and polarise the TME in favour of anti-tumour immunity. Collectively, these three engineered cell types reinforce one another. CAR-NKs and CAR-Ms condition the TME to support deeper CAR-T cell penetration and persistence. Immune-checkpoint receptor knockouts further sustain activity by resisting TME-induced inhibition. This tripartite approach amplifies immune activation, promotes tumour clearance, and represents a promising avenue to overcome the immunosuppressive nature of solid tumours