Tonic signaling of CAR, such as spontaneous CAR activation in the absence of tumor antigen stimulation, can promote T cell exhaustion (88). One approach to silencing tonic CAR signaling is to temporarily halt CAR expression when not needed. For instance, there has been a noteworthy report on a system in which a destabilizing domain (DD) was integrated into the CAR construct to enable the drug-dependent control of CAR protein levels (89, 90). In this setup, CAR expression can be terminated by degradation or restored by the administration of a stabilizing agent for DD to prevent CAR degradation (89) (Figure 1). This system also serves as a safety measure to manage CAR-T cell-related adverse effects because CAR expression can be controlled by ceasing drug administration (90). Inhibition of the epigenetic modifier EZH2 impairs the reversal of CAR−T cell exhaustion, highlighting its role in shaping the exhaustion-associated epigenetic landscape during tonic CAR signaling. Conversely, the clinically available kinase inhibitor dasatinib can transiently block proximal CAR signaling and effectively rejuvenate exhausted CAR-T cells (89). Therefore, the administration of dasatinib during ex vivo expansion of CAR-T cells could represent an additional promising approach for restoring exhausted CAR-T cells without altering the CAR construct. However, while in this preclinical study dasatinib has been shown to modulate proximal CAR signaling and reduce activation/exhaustion programs during production, its effects on in vivo toxicity and clinical tolerability may be context dependent. Indeed, a recent report on CD123-CAR-T cells manufactured in the presence of dasatinib (91) for the treatment of pediatric patients with recurrent/refractory leukemia, showed high-grade cytokine release syndrome (CRS) and immune effector cell-associated hemophagocytic syndrome (IEC-HS) and did not demonstrate enhanced clinical benefit compared to expectations from conventional manufacturing protocols. Possible explanations for the heightened inflammatory toxicity observed include differential antigenic density (e.g., CD123 expression on both malignant and healthy hematopoietic cells), CAR design, disease context (AML/myeloid leukemias vs lymphoid malignancies), and host immune milieu. Therefore, the safety profile of dasatinib exposure during CAR-T cell manufacturing requires careful optimization of dosing, timing, and clinical indication, as well as prospective evaluation of safety outcomes across different CAR targets and patient populations.

In addition to self-activation through tonic signaling, various cellular and environmental factors play a significant role in CAR-T cell exhaustion. Several strategies aimed at preventing exhaustion are presented and discussed below:

Several studies have demonstrated that immune checkpoint blockade is one of the most used and successful strategies to avoid T cell exhaustion (92–94). However, the use of immune checkpoint inhibitors such as anti-PD-1, anti-PD-L1, and anti-CTLA-4 antibodies, independently or in combination with CAR-T cell therapy, has been relatively successful in patients with hematological malignancies (94). Several strategies have been developed to suppress PD-1 function on CAR-T cells, including the combination of anti-PD-1/anti-PD-L1 antibodies (Figure 2A), engineering of CAR-T cells secreting anti-PD-1 or anti-PD-L1 antibodies (Figure 3), or PD-1 gene silencing (Figure 4) (2, 95). Interestingly, CAR-T cells that can secrete anti-PD-L1 or anti-PD-1 antibodies demonstrated higher cytotoxicity and tumor-killing capacity, as well as longer persistence in vivo (96, 97). Recently, PD-1 silenced CAR-T cells have been designed and have shown enhanced anti-tumor effects in vitro and in different cancer mouse models (95, 98). In this strategy, modifying CAR-T cells using gene silencing technology to block the PD-L1/PD-1 immunosuppression axis mainly involves short hairpin RNA (shRNA) or small interfering RNA (siRNA) gene silencing (99, 100) and CRISPR/Cas9 (clustered regularly interspaced short palindromic repeats/caspase 9) gene-editing technology (101, 102) (Figure 4). Interestingly, in a subcutaneous leukemia xenograft, shRNA-mediated gene silencing technology was used to block the effect of PD-1 on the proliferation and anti-tumor effect of CAR-T cells, thereby enhancing its therapeutic effect. In this study, T cells were transduced with the PD-1-shRNA-CAR plasmid using lentivirus to obtain CAR-T cells with PD-1 silenced function (103). The results showed that the efficient PD-1 silencing significantly prolonged the activation and duration of CAR-T cells, resulting in a long tumor-killing effect (103). Building on this strategy of immune checkpoint modulation, anti-CTLA-4 therapy has been shown to enhance T cell priming by disrupting the inhibitory interaction between CD80/86 and CTLA-4, as well as depleting regulatory T cells (Tregs) through antibody-dependent cellular cytotoxicity (ADCC) and phagocytosis (ADCP) (104) (Figure 2B). To further potentiate CAR-T cell function, additional checkpoints, such as CTLA-4, TIM-3, and LAG-3, have also been targeted using gene editing technologies, thereby overcoming multiple layers of immune suppression and enhancing anti-tumor efficacy (105). Interestingly, a double knockout of PD-1 and CTLA-4 using the CRISPR/Cas9 system effectively improved the anti-tumor effects of CAR-T cells (98).

These preclinical data demonstrate that the blockade of T cell inhibitory receptors is a key factor in enhancing the therapeutic effects of CAR-T cells.

Myeloid-derived suppressor cells (MDSCs) represent a critical component of the immunosuppressive tumor microenvironment and pose a significant barrier to CAR-T cell therapy efficacy (106). MDSCs impaired CAR-T cell proliferation and cytotoxic function through several interconnected mechanisms (Figure 5A). One major pathway involves immune checkpoint signaling, wherein MDSCs express PD-L1, which binds to PD-1 on CAR-T cells, inducing T cell exhaustion and diminishing activation (107). In parallel, MDSCs promote the expansion of Tregs, a subset of immunosuppressive lymphocytes that inhibits effector T cell function and proliferation. This is mediated through the secretion of cytokines such as IL-10 and TGF-β (108). These Tregs further suppress CAR-T cell activity by releasing IL-10, IL-35, and TGF-β (109). Additionally, MDSCs contribute to metabolic suppression by depleting essential nutrients, such as cysteine and l-arginine, and producing inhibitory metabolites, such as nitric oxide (NO) and reactive oxygen species (ROS), collectively impairing CAR-T cell viability and adversely affecting persistence (110).

To overcome these immunosuppressive constraints, several strategies aimed at inhibiting Treg and MDSC activity, in conjunction with CAR-T cell therapy, have demonstrated significantly enhanced antitumor responses in various preclinical cancer models. In fact, hampering Treg activity in combination with CAR-T cell therapy has been shown to significantly enhance antitumor responses in various preclinical models of hematologic and solid malignancies (111). Building on this, Long et al. demonstrated that combining GD2-targeted CAR-T cells with strategies to neutralize the immunosuppressive effects of myeloid-derived suppressor cells (MDSCs) markedly improved CAR-T cell efficacy in a sarcoma xenograft model (112). Based on these findings, pharmacological approaches, such as the use of immunotoxins, such as gemtuzumab ozogamicin, have proven effective in depleting MDSCs, thereby further augmenting CAR-T cell responses across multiple tumor types (113) (Figure 5B). Hence, targeting these immunosuppressive cells presents an additional promising avenue to enhance the efficacy of CAR-T cell therapy. Moreover, inhibition of immunosuppressive factors such as IL-10, IL-35, and TGF-β can lead to significant upregulation of T cell memory-related genes and downregulation of exhausted genes (114) [Figures 5A, B]. For example, lenalidomide has been shown to reverse the T cell exhaustion by inhibiting IL-10-induced STAT3 (signal transducer and activator of transcription 3) phosphorylation in patients with CLL (115). A preclinical study demonstrated that silencing endogenous TGF-β receptor II (TGFBR2) in CAR-T cells using CRISPR/Cas9 technology could reduce Treg conversion, prevent CAR-T cell exhaustion, and enhance tumor-killing efficacy in xenograft cancer models (116).

Exhaustion and durable persistence in chronically stimulated T cells (including tumor-infiltrating CAR-T cells) are controlled by interconnected transcriptional circuits that govern lineage progression, effector competence, and epigenetic stability. In this context, TOX/TOX2 and NR4A family members are widely implicated in sustaining exhaustion phenotypes (117, 118). Accordingly, genetic disruption of the TOX/NR4A axis has been shown to reduce dysfunction and improve antitumor activity of CAR-T cells in solid tumor models, including enhanced tumor control with TOX/TOX2 double deficiency and improved efficacy with NR4A ablation (119–124). Recently, it has been shown that the hematopoietic progenitor kinase 1 (HPK1) -NFκB-Blimp1 axis mediates T cell dysfunction, and that high expression of MAP4K1 (which encodes HPK1) correlates with worse patient survival in several cancer types (125). This study suggests that HPK1 is an attractive target for improving the response to CAR-T cell therapy (125). Moreover, overexpression of the transcription factors BATF and IRF4 cooperates to counter exhaustion in tumor-infiltrating CAR-T cells (126). In contrast, TCF1 has been established as a hallmark and functional regulator of progenitor/stem-like exhausted CD8 T cells that retain proliferative potential, sustain a self-renewing pool, and can give rise to more terminally exhausted state. This feature underlies durable responses and responsiveness to immunomodulatory strategies in chronic contexts (127, 128). Consistent with this model, pathways that preserve or restore TCF1-linked programs can enhance persistence: for example, Regnase−1 has been reported to restrain TCF1-associated memory/stem-like features, and its disruption can augment TCF1⁺ CAR-T persistence and antitumor function in relevant settings (117). In ALL, Regnase−1 knockout enhanced TCF1⁺ CAR−T cell persistence and antitumor activity (129). Accordingly, strategies that preserve or enhance TCF1-associated programs can support long-term persistence and functional plasticity of CAR-T cells, especially under chronic antigen exposure.

Similarly, BACH2 functions as a lineage/fate regulator, promoting memory-type programs and mitigating tonic signal–induced dysfunction during CAR-T cell manufacturing (130, 131). However, sustained or dysregulated BACH2 expression may constrain effector differentiation, implying that dynamic tuning rather than constitutive gain or loss may be required for optimal long-term efficacy across CAR constructs and disease contexts. Additional transcriptional strategies that enhance functional resilience include reinforcing AP-1 network balance (e.g. c-Jun) which has been shown to enhance expansion potential, increase functional capacity, diminish terminal differentiation, and improve anti-tumor potency in multiple different preclinical models (132).

These findings highlight that effective manipulation of transcriptional programs in CAR-T cells requires a nuanced understanding of developmental hierarchies and context-dependent function: some factors (e.g., TOX/NR4A) contribute to dysfunctional programs that can be selectively targeted, while others (e.g., TCF1, BACH2) play dual roles in maintaining progenitor potential and influencing effector fate.

Beyond transcription factors, T cell functions are also modulated by metabolic changes. The next section explores metabolic reprogramming and its impact on T cell function and fitness.

The interplay between T cell metabolism and epigenetic regulation is a major area of active research. Metabolic changes can influence the epigenetic landscape of T cells, thereby affecting gene expression and functional outcomes (144). Indeed, mitochondrial membrane potential has been linked to the stemness and longevity of T cells, suggesting that metabolic states can have lasting effects on T cell differentiation and memory (145). In addition, mitochondrial dynamics, including fission and fusion, are essential for maintaining mitochondrial quality control and adapting to changing cellular conditions. Studies have shown that mitochondrial dynamics influence T cell differentiation into effector and memory subsets. For example, effector T cells responsible for immediate pathogen clearance exhibit fragmented mitochondria, favoring glycolysis as the primary energy source (146). Conversely, memory T cells, which provide long-term immunity, display elongated mitochondria and rely on OXPHOS (147). Mitochondrial dysfunction, marked by reduced OXPHOS, increased reactive oxygen species (ROS) production, and altered mitochondrial dynamics, has been implicated in T cell exhaustion (148). In addition, accumulation of damaged mitochondria and impaired mitophagy contributes to a decline in mitochondrial fitness (149).

Mitochondrial fitness, reflected by parameters such as spare respiratory capacity (SRC) and mitochondrial membrane potential (ΔΨm), is critical for sustaining T cell activity under stress conditions, such as during an immune response (150). A high SRC provides a reserve capacity that allows T cells to meet increased energy demands during periods of heightened activity, whereas a high ΔΨm is associated with increased expression of exhaustion markers and higher levels of ROS (145, 150) (Figure 6).

In CAR-T therapy, mitochondrial fitness plays a crucial role in determining the long-term persistence and efficacy of infused T cells (151). Impaired mitochondrial function, characterized by low SRC and increased ROS production, has been linked to T cell exhaustion and reduced therapeutic efficacy (152). This is particularly relevant in diseases such as chronic lymphocytic leukemia (CLL), in which prolonged exposure to tumor cells can impair mitochondrial function, leading to a reduced ability of T cells to mount an effective anti-tumor response (152). In addition, the co-stimulatory molecule 4-1BB promotes mitochondrial fitness and memory-like differentiation (142).

Mitochondrial fitness is pivotal for next-generation immunotherapies aimed at restoring T cell functionality. Targeting mitochondrial biogenesis and ROS regulation may enhance T cell differentiation, activation, persistence, and resistance to exhaustion, offering therapeutic benefits in chronic infections and cancer (153).

Metabolic interventions can improve T cell fitness and persistence of CAR-T cells, thereby enhancing their efficacy in hematological cancer treatment (154). Metabolic limitations may significantly impair T cell physiology, particularly in the context of CAR-T cell therapies (15). Therefore, strategies aimed at enhancing T cell metabolism are critical for improving the efficacy of CAR-T therapies and their anti-tumor activity.

One promising approach involves modulation of signaling pathways that regulate T cell metabolism. However, overexpression of FOXO1 has been shown to enhance CAR-T cell polyfunctionality and metabolic fitness, leading to improved antitumor efficacy without significant toxicity (155). Similarly, co-stimulatory CD28 has been shown to improve T cell glycolytic metabolism, while receptor 4-1BB has been demonstrated to induce mitochondrial biogenesis and enhance metabolic function in T cells, thereby supporting their survival and activity in the TME (156). Thus, engineered CAR-T cells expressing either CD28 or 4-1BB co-stimulatory domains show high efficacy in patients with relapsed hematological malignancies (142).

Additionally, manipulation of cytokine signaling has been explored to enhance CAR-T cell metabolism. Transgenic expression of IL-7 regulates CAR-T cell metabolism and improves persistence in vivo (157). In addition, IL-15 preconditioning has been reported to augment CAR-T cell responses, promoting a stem-like phenotype associated with enhanced engraftment and antitumor activity (158). Interestingly, IL-10-expressing CAR-T cells have shown less T cell exhaustion in the TME through sustained mitochondrial fitness and increased OXPHOS, resulting in the enhanced proliferative capacity and effector function of CAR-T cells (159).

Moreover, the reprogramming of lipid metabolism has emerged as a novel strategy to prevent T cell senescence and enhance therapeutic efficacy. An interesting report demonstrated that inhibiting Cytosolic phospholipase A₂ (cPLA2α) can reprogram lipid metabolism in T cells, resulting in improved antitumor immunity (160). Moreover, it has been reported that a low-arginine microenvironment impairs CAR-T cell proliferation, limiting its efficacy in clinical trials against hematological and solid malignancies (161, 162). To overcome this limitation, modified CAR-T cells expressing argininosuccinate synthase (ASS) and ornithine transcarbamylase (OTC) have been designed to increase proliferation and promote tumor clearance without influencing CAR-T cell cytotoxicity (162). Moreover, CAR-T cells with high glycolytic metabolism and low terminal differentiation demonstrated better CAR-T product outcomes (69). By enhancing the metabolic fitness of T cells, these interventions can potentially overcome the suppressive effects of the TME and boost CAR-T cell efficacy. Moreover, strategies such as increasing mitochondrial spare respiratory capacity (SRC) and reducing ROS production are being explored for their potential to enhance CAR-T cell performance (163). These approaches may involve the use of drugs that promote mitochondrial biogenesis as well as inhibitors targeting tumor-specific metabolic pathways, reducing competition for energy resources between tumor cells and CAR-T cells (164). In addition, activation of peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α), a key regulator of mitochondrial biogenesis, has been shown to improve the mitochondrial fitness of T cells and enhance their anti-tumor activity (165, 166).

Metabolic and mitochondrial fitness act as the energetic backbone of T cell fitness, enabling sustained signaling, resistance to exhaustion-associated transcriptional programs, and survival within nutrient-restricted tumor microenvironments.

In summary, leveraging metabolic reprogramming offers a promising approach for enhancing T cell fitness in CAR-T cell therapy (Table 1). Future research should focus on integrating these metabolic strategies into clinical protocols to maximize the therapeutic potential of CAR-T therapies.

Optimal T cell fitness ensures a rapid and robust immune response to infections, cancer, and other immune challenges, and plays a vital role in overall immune health and disease defense. In contrast, chronic inflammation may be perceived as a state of prolonged and persistent inflammatory response that occurs when the immune system continues to attack perceived threats long after the initial injury or infection has resolved (181). Interestingly, chronic inflammation and T cell fitness are closely interconnected, with each influencing the other (182). Chronic inflammation can lead to the continuous activation of T cells, gradually impairing their fitness, affecting their function, promoting tumor development, and suppressing anti-tumor immunity. In addition, T cells with memory-like phenotypes are known to retain higher fitness, underscoring the impact of the host inflammatory environment on T cell quality and therapeutic potential (183). Indeed, prolonged exposure to inflammatory signals can cause T cell exhaustion and consequently limit their ability to effectively control infections and tumors (184). For example, in chronic inflammatory settings such as CLL, memory-like T cells with lower exhaustion exhibit superior function, whereas exhausted, inflammation-driven T cells with metabolic dysregulation display reduced fitness and are associated with poorer responses (69, 185). Further, chronic inflammation can contribute to a decline in the diversity of the naive T cell repertoire, limiting the immune system’s ability to recognize and respond to novel antigens (186). This decline is particularly pronounced with aging, and can increase susceptibility to infections and reduce vaccine efficacy (187). In particular, senescence-associated T cells (SA-T cells) secrete proinflammatory factors that contribute to chronic inflammation and autoimmunity (188). In this context, inflammatory mediators are known to accelerate T cell senescence, causing T cells to lose their functional capacity and contributing to the overall inflammatory milieu (189). Several inflammatory mediators play key roles in modulating T cell fitness during chronic inflammation. Indeed, pro-inflammatory cytokines such as tumor necrosis factor alpha (TNF-α) and IL-6 have been shown to alter T cell signaling thresholds, leading to impaired activation and function, particularly in the context of aging (184) (189). Furthermore, eicosanoids, which are inflammatory lipids such as prostaglandins, can promote inflammation and affect T cell function (190).

(190). Therefore, understanding the mechanisms by which inflammation affects T cells is crucial to develop strategies to mitigate these effects and improve immune health.

Aging contributes to immunosenescence, marked by a decline in naïve T cell production and diminished T cell diversity (191). As T cells age, they exhibit reduced proliferative capacity, impaired cytokine production, and increased expression of inhibitory receptors, all of which contribute to T cell exhaustion and decrease the effectiveness of CAR-T therapy in older patients (191). Furthermore, aging significantly compromises T cell fitness, a decline shaped by various patient-specific factors. Therefore, in older patients, CAR-T cell therapy efficacy can be compromised by poor fitness of the starting T cell material (192).

Mitochondrial dysfunction is another consequence of aging that further impairs the fitness and functionality of T cells, exacerbating the challenges in achieving optimal CAR-T cell efficacy in elderly individuals (192).

T cell senescence leads to loss of proliferative capacity and increased accumulation of dysfunctional CD4+ T cells, termed senescence-associated T (SA-T) cells (188).

Thus, efficient metabolic programming is crucial for T cell homoeostasis and differentiation (193). Age-related metabolic dysregulation can impair T cell responses, particularly in the context of infections and vaccinations (193). Moreover, genome-wide analyses have revealed significant transcriptional changes in aged T cells, affecting gene expression and signaling pathways crucial for T cell function (194).

While these factors highlight the challenges of T cell aging, ongoing research aims to develop interventions that could rejuvenate T cell function, potentially improving the outcomes of immunotherapies such as CAR-T cell therapy for older patients (192).

The type of cancer being treated significantly influences the fitness of T cells used in CAR-T therapy (195). Different cancers have distinct tumor microenvironments that can either support or hinder CAR-T cell function. Notably, clinically relevant differences in baseline T cell differentiation, exhaustion status, and inflammatory exposure have been consistently reported across ALL, NHL, MM, and CLL, which has direct implications for manufacturing success, in vivo persistence, and therapeutic durability (196, 197). This is partly due to differences in the immune evasion strategies employed by these tumors as well as the cumulative impact of prior therapies on the patient’s immune system.

In ALL, T cells tend to retain a higher proportion of naïve and early memory subsets, which is associated with better expansion and persistence after CAR-T infusion (198). This phenotype is consistent with the high response rates and long-term persistence observed in CD19-directed CAR-T trials in pediatric and young adult ALL, representing one of the most clinically favorable settings for CAR-T therapy. In contrast, T cells from patients with NHL and MM often show signs of chronic activation and exhaustion, including upregulation of inhibitory receptors and a shift toward more differentiated effector phenotypes (199). These disease contexts are therefore characterized by intermediate CAR-T fitness constraints, where clinical strategies such as optimized leukapheresis timing, selection of less-differentiated T cell subsets, and improved culture conditions are increasingly implemented, while pharmacologic or genetic rejuvenation approaches largely remain preclinical. This makes it more challenging to achieve sustained responses in these patients, necessitating additional interventions to rejuvenate or enhance the fitness of collected T cells.

As previously mentioned, age-related inflammation can contribute to T cell dysfunction, and the presence of cancer may intensify inflammation, further accelerating T cell aging (200).

Numerous hematological malignancies, such as MM and CLL, show elevated levels of inflammatory cytokines, such as IL-6, both locally and systemically, which correlates with worse outcomes (201). MM, in particular, relies heavily on IL-6 during certain phases of its progression (201), and the presence of increased levels of the inflammatory cytokine IL-17 or T helper 17 cells in peripheral blood mononuclear cells (PBMCs) and the bone marrow environment is linked to worsened disease (202). In relapsed/refractory (R/R) MM, inflammation from the disease enhances cytokine signaling pathways and proliferation driven by lymphopenia, thereby accelerating immune senescence (203). This leads to a decrease in naive T cells and changes in the metabolism of the remaining T cells in patients with MM.

Moreover, research conducted by Suen and colleagues identified that T cell clones from MM patients displayed a senescent effector phenotype, underscoring the T cell dysfunction associated with MM. Clinically, these features translate into reduced CAR-T persistence and increased risk of early functional decline, motivating the exploration of manufacturing-stage interventions (e.g., cytokine modulation, signaling inhibitors, metabolic reprogramming), which are currently under early clinical or advanced preclinical investigation.

Similarly, CLL is associated with early T cell aging (204) and represents one of the most challenging indications from a T cell fitness perspective. Patients with CLL typically exhibit a low CD4:CD8 ratio, which correlates with a shorter overall survival rate (204). Additionally, T cells from these patients show inherent functional impairments; when stimulated in vitro, their proliferation is significantly lower than that of T cells from age-matched R/R MM patients or young adults with ALL (205). Although CD8 T cells in patients with CLL can produce cytokines such as IFNγ and TNF, their cytotoxic function is hindered by issues such as improper granzyme localization and higher levels of PD-1 and other inhibitory receptors (64). This suggests that CLL contributes to a state of exhaustion in which T cells can recognize tumor antigens but are unable to effectively manage the disease. From a therapeutic modulation standpoint, most strategies in CLL including checkpoint blockade, cytokine pathway inhibition, epigenetic modulation, and signaling “rest” approaches remain largely preclinical or investigational, with limited clinical validation to date.

Acute myeloid leukemia (AML) represents a clinically important and biologically distinct context in which T cell fitness constitutes a profound limiting factor (206, 207). Although several early-phase CAR-T trials are ongoing in AML, no CAR-T therapy has yet been approved for clinical use in this indication, underscoring the substantial barriers to successful translation (208). In AML, T cells frequently exhibit baseline features of advanced differentiation, functional exhaustion, and metabolic impairment, which undermine proliferative capacity and persistence even prior to CAR engineering (209, 210). Clinical failures/limited efficacy of immune checkpoint blockade in AML (211, 212) further illustrates that simple reversal of PD-1/PD-L1 signaling is insufficient to restore T cell function in this disease. This limited clinical benefit reflects the fact that Tcell dysfunction in AML is not driven by checkpoint signaling alone but instead arises from multilayered immunosuppressive mechanisms operating within the leukemic niche. Indeed, the AML tumor microenvironment is highly immunosuppressive, characterized by abundant Tregs, MDSCs, and suppressive metabolites (e.g., adenosine, kynurenine) that directly impair effector T cell activation, cytokine production, and proliferation (213, 214). Additionally, AML blasts exhibit downregulation of antigen presentation machinery and heterogeneous expression of target antigens, which limit both endogenous and engineered T cell targeting efficacy and promotes immune escape (215).

Together, these intrinsic and extrinsic constraints place AML among the most challenging hematological malignancies for CAR-T therapy, and most approaches to enhance T cell fitness in AML remain preclinical or investigational, including metabolic reprogramming, dual-antigen targeting, adaptation of culture conditions, and modulation of niche-derived suppressive signals (208, 216). This disease thus exemplifies the multifaceted barriers to CAR-T efficacy that extend beyond T cell exhaustion alone and illustrates the need for disease-specific therapeutic modulation strategies.

Collectively, these findings illustrate that cancer type and its associated inflammatory and treatment context impose distinct, disease-specific barriers to CAR-T efficacy, ranging relatively mild constraints in ALL to profound intrinsic dysfunction in CLL, advanced MM, and the highly immunosuppressive setting of AML. Systematic integration of clinically established optimization strategies with emerging preclinical T cell rejuvenation approaches will therefore be essential to tailor CAR-T manufacturing and treatment paradigms to each disease setting. A comparative overview of disease-specific CAR-T fitness constraints and modulation strategies is provided in Table 2.

In addition to aging and cancer types, treatments can contribute to T cell dysfunction in patients with cancer. The phenotype and proliferative capacity of T cells in patients with hematological malignancies are also affected by prior treatment. Some therapies can harm the T cell pathways involved in the repair of DNA damage or protein turnover. Particularly, significant lymphopenia caused by drug or radiation exposure is considered a major factor contributing to T cell aging (217).

ASCT is a key therapy for hematological cancers. However, it perturbs the composition of the T cell compartment and causes significant metabolic changes, ultimately reducing T cell fitness (203). Changes in T cell composition caused by ASCT and other conventional therapies that induce lymphopenia are irreversible, resulting in permanent increases in central memory T cells and effector memory T cells in patients with MM (203). Moreover, the CD4/CD8 ratios can be affected by this treatment due to the increased proliferation of CD8 T cells in comparison to CD4 T cells in the lymphopenic environment following transplantation. In children and young adults receiving ASCT, the CD4:CD8 ratio typically returns to normal levels within a year after treatment (218), whereas in older adults, this ratio may take longer to normalize, if it ever does (219). Moreover, it has been shown that patients with B-cell malignancies typically have a lower CD4/CD8 ratio, fewer naïve T cells, and a more differentiated T cell phenotype that worsens after multiple chemotherapy lines (220, 221). In a recent preclinical study, Das et al. showed that cumulative chemotherapy cycles deplete naïve T cells in many pediatric cancer regimens, reducing the expansion potential of CAR-T cell therapy (221). In contrast, the chemotherapeutic drug bendamustine can lead to prolonged lymphopenia, increased Tregs, and enhanced function of MDSCs, negatively impacting CAR-T cell production and response in aggressive lymphomas (222–224). In MM, exposure to alkylating agents promotes more senescent CD8+ T cells, whereas immunomodulatory drugs improve T cell function (225). With the increase in immunotherapeutic options for the treatment of hematological malignancies, the ideal order for using such therapies, particularly when they target the same antigen, is still unclear and needs to be taken into consideration before any therapeutic decision. For example, patients with MM who experience disease progression after BCMA-directed therapies can still respond to BCMA-CAR-T cell treatment, but response rates and durability were suboptimal compared to those not treated with prior BCMA-directed therapies (226, 227). It has been previously shown that continuous exposure to bispecific antibodies, such as blinatumomab, which stimulates T cells via the TCR but lacks a co-stimulatory signal, may result in T cell exhaustion or anergy, potentially reducing the effectiveness of later CAR-T cell therapy (228). However, treatment-free intervals may help restore T cell phenotype and function (228). Studies in pediatric B-ALL showed that blinatumomab exposure is linked to a higher risk of early treatment failure and reduced CD19 expression in non-responders (229–231). Post-CAR consolidation with hematopoietic stem cell transplantation may be a beneficial approach to mitigate the risk of relapse in blinatumomab-exposed patients. Interestingly, the use of CD19 immunotherapies, such as tafasitamab and loncastuximab, prior to CAR-T cell treatment in patients with DLBCL does not preclude subsequent responses to CD19-directed CAR-T cell therapy (232, 233). Owing to their long half-life and possible competition for CD19 binding with CAR-T cells, a washout period may be advisable. Nevertheless, CAR-T cell therapy remains more effective than chemotherapy in patients with MM who have undergone extensive prior treatment and progress after receiving a bispecific antibody (234). Therefore, it is critical for successful CAR-T cell therapy to select treatments that can positively affect T cell fitness. For example, Bruton’s tyrosine kinase inhibitor (BTKi) ibrutinib administered to CLL patients during apheresis positively affects CAR-T cell response by reducing the immunosuppressive environment, enhancing TCM proportion, and T cell expansion, resulting in significant complete response (CR) rates (235). However, the timing of ibrutinib administration relative to leukapheresis is critical because its effect may vary based on the duration before or after treatment (236, 237). In r/r Mantle Cell Lymphoma (CML), the TARMAC trial (NCT04234061) showed promising results with time-limited ibrutinib (≥7 days before leukapheresis until MRD−) in combination with CAR-T therapy (238).

Moreover, some therapies, such as immunomodulatory drugs, have the potential to boost CAR-T cell efficacy after infusion and could be a promising approach in the treatment of hematological malignancies. In LBCL patients who relapsed after CAR-T cell treatment, lenalidomide administration showed improved overall survival compared to chemotherapy (239). Similarly, in MM, the treatment of patients with lenalidomide or pomalidomide post-CAR-T cell therapy has been shown to be safe and leads to CAR-T cell re-expansion, late-onset, and durable clinical responses in some patients (240).

In summary, integrating assessments of CAR-T cell activity and longevity along with monitoring residual disease can assist in forecasting disease relapses, enabling proactive measures before visible recurrence occurs.

Considering the individual effects of aging, chronic infections, cancer type, and treatment on T cell fitness, the cumulative impact of these factors on T cell fitness, particularly in older heavily treated patients, needs to be taken into consideration. Therefore, refined measures should be developed to assess T cell fitness to allow for a better design of T cell-based immunotherapeutics for older and heavily treated patients with hematological cancers contemplating cell therapies.

The manufacturing process of CAR-T cells is complex and subject to numerous variations at each step. These variabilities, such as the time of T cell collection and treatment, can affect the quality of the final product, ultimately impacting clinical outcomes. The timing of T cell collection, whether early in the disease course or after multiple lines of therapy, plays a significant role in determining the quality of the CAR-T product.

Early collected T cells tend to have a higher proportion of less differentiated subsets, such as TN and TSCM cells, which are crucial for the long-term persistence and efficacy of CAR-T cells (198). In contrast, T cells collected late in the disease course, after multiple rounds of chemotherapy, or in the setting of advanced disease, are often more differentiated, exhibit higher levels of exhaustion markers, and have reduced proliferative capacity (241). It has been shown that T cells collected at first relapse led to higher naïve T cell percentages and lower exhaustion markers, resulting in improved anti-CD19 CAR-T cell fitness in patients with r/r LBCL, showing a better overall response rate (ORR) (242). Similarly, in MM, T cells collected from patients early in myeloma therapy exhibit better fitness for CAR-T manufacturing, with a higher proportion of early memory CD8+ T cells, a better CD4/CD8 ratio, and greater proliferative capacity compared to those who were relapsed or refractory (243).

Moreover, it has been demonstrated that intensive previous chemotherapy lines are linked to a reduced response to CAR-T cell treatment in LBCL (244).

These observations underscore the importance of considering the timing of T cell collection as a key factor in optimizing the fitness of CAR-T products. In clinical practice, this may involve proactive monitoring of patients at high risk of relapse and initiating T cell collection before disease progression or before administering therapies that could compromise T cell fitness.

Table 3 summarizes the key patient-specific factors that critically impact T cell fitness and, consequently, the efficacy of CAR-T cell therapy.