Although high TIL levels generally predict better therapeutic outcomes (9), immune-excluded tumors exhibit poor responses due to T cells being trapped in the tumor stroma, unable to penetrate the tumor core. This phenomenon, often mediated by CAFs and TECs, results in limited T cell-tumor contact and suboptimal ICB efficacy (10-12). CAFs, originating from diverse sources such as mesenchymal cells, epithelial-mesenchymal transition and stromal activation, exhibit functional plasticity (13). CAFs polarize into inflammatory CAFs, which secrete immunosuppressive cytokines, or myofibroblastic CAFs, which create physical barriers through ECM remodeling (14). A small subset of CD74⁺ CAFs can also present antigens, directly modulating immune responses (15).

Similarly, TECs impede T-cell infiltration through abnormal vascular responses, including reduced adhesion molecule expression. Abnormal responses of blood vessel TECs to inflammatory cytokines and decreased expression of adhesion molecules impede T-cell recruitment and infiltration (16,17). Studies have demonstrated that high inflammation coupled with low vasculature correlates with optimal ICB responses, as confirmed by two independent investigations (18,19). Lymphatic TECs further exacerbate immune exclusion by attracting C-X-C chemokine receptor type 4 (CXCR4)⁺ T cells away from tumors via C-X-C motif chemokine ligand 12 (CXCL12) signaling (20). Collectively, these stromal cells establish chronic inflammatory niches that hinder effective immune responses. Recently, research (21) in hepatocellular carcinoma has revealed that pericancerous macrophages cross-present antigens to CD103⁺ CTLs via the endoplasmic reticulum-associated degradation machinery-mediated cytosolic pathway, subsequently activating the NLRP3 inflammasome in macrophages, thereby promoting the progression of hepatoma and immunotherapy resistance.

T-cell specificity, driven by T-cell receptor (TCR) recognition of major histocompatibility complex-I:antigen complexes, underpins T-cell activation and antitumor immunity. TMB and neoantigen availability strongly correlate with ICB efficacy in immunologically 'hot' tumors such as melanoma and lung cancer (22-24). Conversely, in 'cold' tumors with low TMB, treatments such as chemotherapy, radiotherapy and tumor vaccines can enhance antigenicity, converting cold tumors into hot ones (25-28).

Despite these advances, most T cells in the TME are bystander T cells (Tbys) that lack tumor antigen recognition but still influence immune responses (29). Studies suggest that expressing Tby-specific antigens in tumor cells could harness their antitumor potential (30-32). By contrast, antigen-specific T cells (Tas), which recognize tumor neoantigens, are the primary effectors in immunotherapy (33). These cells often exhibit exhaustion due to chronic stimulation (34), characterized by markers such as CD39 (31,35,36,37), CD103 (38-40) and programmed cell death protein 1 (PD-1) (41,42). Emerging markers such as CXCL13 (43-47) and CD137 (48,49) further refine Tas identification and therapeutic targeting. Notably, the predominant expression of exhaustion markers among Tas markers emphasizes the critical need to address T-cell exhaustion in therapeutic strategies to maintain clonal expansion and effector function (50).

Chronic antigen exposure in tumors induces T-cell exhaustion, marked by impaired proliferation, reduced cytokine production and elevated inhibitory receptor expression (51,52). T-cell exhaustion represents a complex process orchestrated by dynamic alterations and dysregulation in signaling pathways, transcription factors [including T cell factor 1 (TCF1), T-box expressed in T cells and thymocyte selection-associated high mobility group box protein (TOX)], epigenetic programs and metabolic adaptations (53,54). Exhausted T cells (Tex) comprise a heterogeneous population exhibiting a spectrum of phenotypic and functional characteristics (55). This spectrum is anchored by two distinct states: Progenitor Tex (Tex-prog), which maintain stem-like and proliferative properties, and terminally differentiated Tex (Tex-term).

Tex-prog cells are characterized by high expression of TCF1 (56,57), a transcription factor essential for maintaining stemness and proliferative capacity. These cells respond well to ICB (58-62) and IL-2 (63-65) therapies. Tex-prog cells are considered primary mediators in chronic infection and ICB response (66-69), serving as primary mediators of long-term immune responses (70). The presence of the cells correlates with improved survival in melanoma and other cancer type. Notably, CXCL13⁺ T cells also exhibit Tex-prog characteristics (71).

Tex-term, defined by TOX activity (72-75) and multiple inhibitory checkpoints [e.g., PD-1 and lymphocyte activation gene 3 (LAG3)] (76-78), exhibit irreversible epigenetic changes, rendering them resistant to ICB (69). Under hypoxic conditions, Tex-term suppress immune responses via CD39-adenosine signaling (79), contributing to immunotherapy resistance. Strategies targeting Tex-term-associated pathways, such as P-selectin glycoprotein ligand-1 (80) and CD39 (81,82), are under investigation.

Tex-int cells occupy an intermediate state between functional effector T cells and Tex-term. These cells retain proliferative capacity and respond robustly to ICB, serving as a 'stem-like' reservoir for sustained immune responses (68,69). The abundance of these cells correlates with improved clinical outcomes, making them promising targets for combination therapies such as vaccines and chimeric antigen receptor (CAR)-T cells (53).

T-cell exhaustion results from persistent antigen stimulation and is regulated by a complex interplay of signaling pathways (71), transcription factors and epigenetic programs. Factors such as mechanistic target of rapamycin (mTOR), transforming growth factor-β (TGF-β) and PD-1 signaling influence transcriptional regulators, such as basic leucine zipper transcription factor ATF-like (BATF), TOX and TCF1, which maintain a delicate balance between T-cell activation and exhaustion (83,84). Briefly, upstream molecular pathways, such as mTOR and TGFβ (85), PD1 signaling (86) and IL-10 stimulation (87), orchestrate the activities of transcription factors myeloblastosis (88), activator protein 1 (61,89), BATF (90-92), BTB and CNC Homology 2 (BACH2) (90), special AT-rich sequence-binding protein 1 (93), interferon regulatory factor 4 and nuclear factor of activated T cells cytoplasmic 1 (91,92,94). Multiple aspects of a T cell are kept in a fine balance by these regulators; for example, methylcytosine dioxygenase TET2 guards against BATF3-induced CAR-T cell proliferation and ensuing genomic instability (95,96). With a more complete understanding of this epigenetic programming, it is possible to fine tune T-cell immunity and exhaustion while keeping proliferative responses and general inflammation in check.

The systemic T-cell response to ICB is driven by two key mechanisms: The expansion of pre-existing T-cell clones and the infiltration of new clonotypes, a process termed clonal rejuvenation (97). Pre-existing clones provide a rapid antitumor response, while new clonotypes, undetectable before treatment, emerge post-ICB therapy (98). This dynamic relies heavily on continuous trafficking (99,100) between the lymphatic system and blood circulation. Notably, blocking lymphocyte egress from lymph nodes with the sphingosine-1-phosphate receptor inhibitor FTY720 abolishes ICB efficacy (101), underscoring the critical role of the peripheral immune system in T-cell activation and expansion.

Peripheral lymphoid structures are central to adaptive immunity, serving as hubs where antigens, antigen-presenting cells (APCs) and lymphocytes interact within specialized microenvironments (Fig. 3). These interactions occur in secondary lymphoid organs (SLOs), such as tumor-draining lymph nodes (tdLNs), and in tertiary lymphoid structures (TLSs), which form in response to chronic inflammation (102). Both SLOs and TLSs facilitate antigen-specific T-cell activation, playing pivotal roles in antitumor immunity (103).

tdLNs are critical for antitumor immunity (104) and serve as both initial metastatic sites and primary locations for T-cell activation. The presence of these lymph nodes correlates with ICB efficacy (105), systemic immune responses (106) and improved survival (107). Mechanistically, tdLNs (108) support antitumor immunity through two pathways: De novo T-cell generation and maintenance of pre-existing T cells. Specialized dendritic cells (DCs) transport tumor antigens from the TME to tdLNs (109), where resident DCs activate naive T cells, initiating a cycle of T-cell trafficking between the lymph nodes and the TME (110). tdLNs harbor TCF1⁺ T cells with superior functionality compared to TILs (111,112). These cells, maintained by conventional type 1 dendritic cells (113,114), expand upon ICB treatment (62,115) and are key drivers of T-cell infiltration into the TME (105).

However, tdLNs can also foster immunosuppressive environments, with PD-L1⁺ (116) myeloid cells and regulatory T cells (Tregs) dampening T-cell responses. Tumor metastasis to tdLNs exacerbates this suppression, promoting distant metastasis and poor prognosis (117,118). Optimizing tdLN function presents a promising avenue for next-generation immunotherapies.

TLSs are organized immune cell aggregates that form in chronic inflammatory conditions, including cancer (119-121). Structurally resembling SLOs, TLSs support de novo and ongoing immune responses, promoting T-cell activation and B-cell-mediated antibody production (122). Their presence is associated with favorable outcomes across multiple cancer types, including breast (123), colorectal (124) and pancreatic (125,126) cancer.

TLS formation is driven by chemokines such as CXCL13, and their maturity, ranging from early lymphocyte aggregation to fully developed germinal centers, impacts their prognostic value (127-129). Mature TLSs are linked to better outcomes, while immature TLSs may correlate with a poor prognosis in certain cancer types, such as lung and breast cancer (130-132). Emerging therapeutic strategies aim to enhance TLS formation (133-136) and maturation while mitigating tumor-promoting inflammation.

The heterogeneity in TLS maturation, categorized into early (lymphocyte aggregation), primary (follicular dendritic cells without mature DCs) and secondary stages [active germinal centers (GCs)] (137), is evident across cancers and assessed via markers such as CD138⁺ plasma cells (138) and CD23⁺ GC B cells (139). For example, esophageal cancer shows 26.9% TLS-negative, 30.2% immature and 42.9% mature cases, with mature TLSs featuring proliferative/memory B cells, plasma cells and CD4⁺ Th17 cells (140). Liver cancer exhibits 53% TLS-negative cases, 26% lymphoid aggregates, 16% primary and 5% secondary follicles, with the latter two linked to reduced recurrence risk (141). In squamous lung cancer, mature TLSs with GCs predict prognosis unless diminished by corticosteroids during neoadjuvant chemotherapy (128,142). Similarly, melanoma highlights B-cell markers (activation-induced cytidine deaminase and CD21) for mature TLSs and prognostic insights (143), while colorectal cancer emphasizes the significance of follicular dendritic cells and mature B cells (144).

However, TLSs may correlate with poor outcomes in some cancer types (145,146), such as renal cell carcinoma and pancreatic ductal adenocarcinoma, potentially due to TLS-associated Tregs (147), IL-10-secreting B cells (148), immune complexes, or tumor-promoting inflammation and angiogenesis (149,150). Emerging therapies aim to foster TLS neogenesis/maturation while mitigating inflammatory TMEs using cytokines, antibodies, immune checkpoint inhibitors, cellular therapies and synthetic scaffolds (151).

Systemic inflammation plays a pivotal role in regulating T-cell responses during immune defense against malignancies. Pro-inflammatory mediators, such as IL-1β, TNF-α and IL-6, create an inflammatory environment that facilitates APC-mediated T-cell activation and expansion. While this process is essential for initiating immune responses against infections and cancer, chronic inflammation can lead to T-cell exhaustion (211), marked by reduced effector functions and increased expression of inhibitory receptors such as PD-1 and CTLA-4.

Therapeutic strategies targeting systemic inflammation aim to enhance T-cell functionality while preventing exhaustion. Approaches include use of cytokine inhibitors (e.g., IL-6 or TNF-α antagonists) to mitigate excessive inflammation and checkpoint inhibitors to restore T-cell activity (212). Emerging therapies such as selective adjuvants that activate Toll-like receptors and metabolic interventions targeting T cell-specific pathways offer promising avenues to sustain T-cell function without inducing chronic inflammation, particularly in cancer immunotherapy contexts (213).

PD-1 blockade therapy has transformed cancer treatment by reinvigorating T cell-mediated antitumor immunity. PD-1, an immune checkpoint expressed on T cells, suppresses T-cell activation to maintain self-tolerance, but is exploited by tumors for immune evasion via the PD-1/PD-L1 axis (214). Monoclonal antibodies targeting this pathway, such as nivolumab and pembrolizumab, have achieved marked clinical success across various cancer types (215-217), such as non-small cell lung cancer, head and neck squamous cell carcinoma, and melanoma. Although these therapies provide durable responses and improved survival, they can cause irAEs, necessitating careful management (218,219). Combining PD-1 inhibitors with other immunotherapies or conventional treatments has shown promise in overcoming resistance and achieving synergistic effects (220-223).

Advances in vaccine technologies have significantly enhanced T cell-mediated antitumor immunity. Cancer vaccines targeting tumor-associated antigens stimulate specific T-cell responses by mimicking natural infections (224-227). The addition of adjuvants, such as CpG oligodeoxynucleotides, further amplifies innate immune responses, supporting T-cell activation and proliferation (228-231). Combining cancer vaccines with checkpoint inhibitors has demonstrated synergistic effects, improving T-cell responses and tumor clearance (225,232,233). Personalized cancer vaccines, tailored to individual tumor antigenic profiles, represent a promising breakthrough by generating broad-spectrum T-cell responses while minimizing immune escape. These developments mark a significant step forward in precision immunotherapy (234).

mRNA-based therapies have emerged as a powerful tool in cancer immunotherapy, leveraging the ability of mRNA to encode tumor-specific antigens and stimulate targeted immune responses. Advances in delivery systems, such as lipid nanoparticles, have improved mRNA stability and protection against degradation (235-237). Structural modifications, including nucleoside alterations, enhance translation efficiency while minimizing immunogenicity (238-241). Co-delivery of immunomodulatory molecules encoded by mRNA further amplifies T-cell responses and counters immunosuppression (242). Clinical trials, particularly in melanoma, have demonstrated the efficacy of personalized mRNA vaccines in eliciting robust T-cell responses, underscoring their potential in cancer treatment (243-245).

Lymphoid tissue-targeted nano delivery systems offer an innovative approach to enhancing T-cell functionality in cancer immunotherapy. Engineered nanoparticles, modified with ligands targeting lymphoid tissue receptors (e.g., CCR7 and CXCR5), enable precise delivery of immunomodulatory agents (246-249). Encapsulation of cytokines, such as IL-2 and IL-12 (250-253), within these nanoparticles enhances local T-cell responses while minimizing systemic side effects. Recent advancements include delivering checkpoint inhibitors directly to lymphoid tissues, addressing T-cell exhaustion in the TME (254-257). By concentrating therapeutic agents at desired sites, these systems improve treatment efficacy and reduce adverse effects, representing a promising direction for cancer immunotherapy.

OVs present a novel strategy to enhance T cell-mediated antitumor responses by selectively infecting and lysing cancer cells. This process triggers pro-inflammatory responses that activate adaptive immunity while sparing healthy tissue (258). OVs can also be engineered to express immunomodulatory genes, such as IL-12 and granulocyte-macrophage colony-stimulating factor, which enhance T-cell recruitment and activation within the TME (259-262). Combining OVs with other immunotherapies, particularly checkpoint inhibitors, has shown synergistic effects in overcoming tumor-induced immunosuppression (263). This dual mechanism, namely direct tumor lysis and immune activation, positions OVs as a promising component of combination immunotherapy strategies.