Immunotherapy is a promising new treatment for cancer, one which harnesses a patient’s own immune system to target tumor cells. These therapies aim to boost anti-tumor activity of cytotoxic immune cells, such as cytotoxic CD8⁺ T cells and Natural Killer (NK) cells. During acute infections, naïve CD8⁺ T cells undergo robust proliferation and clonal expansion to differentiate into effector and memory CD8⁺ T cells. In contrast, in cancer and during chronic infections persistent antigen stimulation abrogates the development of memory T cells and T cells become exhausted. In cancer, immunotherapies attempt to enhance CD8⁺ T cell effector function by targeting various inhibitory receptors on the cells’ surfaces. For example, anti-programmed death protein 1 (PD-1) monoclonal antibodies are exceptional for PD-1 receptor inhibition, thereby promoting anti-tumor CD8⁺ T cell activity (Hellmann et al., 2019; Hellmann and Ramalingam, 2020). Anti-PD-1 therapies such as pembrolizumab and nivolumab were first approved for melanoma (Wolchok et al., 2013) and have since shown exceptional results in other cancers as well (Tawbi et al., 2018; Doki et al., 2022). Similarly, chimeric antigen receptor-T (CAR-T) cell therapy harnesses T cell-mediated anti-tumoral immune activity, an immune therapy approach that optimizes T cell activity against tumor cells. As promising as this therapy is for specific types of hematological malignancies, CAR-T therapy are less effective against solid tumors, and are also associated with some significant safety concerns (Gust et al., 2020; Rubin and Vaitkevicius, 2021). Specifically, in both CAR-T and immune checkpoint blockade (ICB) such as anti-PD-1 therapy, resistance to treatment remains a significant issue that limits their efficacy and utilization for a large subset of cancer patients (Gust et al., 2020; Rubin and Vaitkevicius, 2021). Therefore, defining the mechanisms that mediate resistance to CTL-based immunotherapies is of upmost importance (Sade-Feldman et al., 2018; Klemen et al., 2019; Lu et al., 2019).

One mechanism of resistance to T cell-targeting immunotherapies is loss of T cell effector activity (Pauken et al., 2016; Balança et al., 2021), whereby dysfunctional/exhausted CD8⁺ T cells fail to adequately clear cancer cells. Recently, epigenetic alterations were found to reinforce the differentiation of naive T cells to the exhausted state (Yang and Wang, 2021; Belk et al., 2022; Chen et al., 2022). Therefore, we and others believe that CD8⁺ T cells can be re-sensitized to ICB therapy by modulating the activity of specific epigenetic factors that mediate CD8⁺ T cell exhaustion. Here, we will describe effector and exhausted CD8⁺ T cell differentiation in the tumor microenvironment, explore transcription factors and genomic alterations reinvigorating such differentiation, and examine therapies which target these epigenetic pathways as a means to enhance the efficacy of ICB therapy.

CTLs are very effective at removing foreign invaders in the body, such as during a viral infection. To do so, CTLs differentiate into their antigen-specific subtypes. Upon antigen recognition, naïve CD8⁺ T cells undergo clonal expansion and differentiate into effector CD8⁺ T cells, as shown in Figure 1. These immune cells are capable of cytokine secretion, have cytolytic activity, and function to target and clear virally infected cells. Once an infection is cleared from the body, the majority of these effector cells die leaving a small percentage (about 5%) of them to differentiate into long-lasting memory T cells (Tonnerre et al., 2021). This process in comparison with chronic antigen exposure is illustrated in Figure 1.

In the setting of a chronic antigen exposure, such as in cancer or chronic viral infections, the effector activity of CD8⁺ T cells is hindered. Even in tumors with a high mutational burden and the existence of a large amount of neoantigens, the immune system fails to completely clear all cancerous cells (Caushi et al., 2021; Belk et al., 2022). Thus, CD8⁺ T cells in the TME are thought of as being dysfunctional throughout tumorigenesis. This is mainly due to persistent antigenic stimulation during chronic infections or in the TME where CD8⁺ T cells lose their effector functions and are skewed towards an exhausted phenotype. There is now a wide belief that terminally exhausted CD8⁺ T cells are attributed to immune checkpoint blockade failures in the clinic, a point which will be discussed later in this review.

T cell differentiation into an exhausted phenotype is a dynamic process, moving through a series of cellular states. Upon consistent antigenic stimulation either during chronic viral infections or cancer, effector CD8⁺ T cells begin to lose their proliferative capacity, cytokine production, and effector potential. They upregulate various inhibitory receptors (such as PD-1, CTLA4, Tim3, LAG3, etc.), alter known transcription factors (such as TCF1, T-bet, NFAT, TOX, Blimp-1, etc.), and incur a unique epigenetic state, all of which hinder their effector and cytolytic abilities (Tang et al., 2021). The upregulation of inhibitory receptors on exhausted CD8⁺ T cells, especially on progenitor exhausted CD8⁺ T cells (Tang et al., 2021), are targeted with ICBs and as a means of stimulating CTL effector functions of exhausted T cells. It is important to note that although chronic viral infection and cancer can both cause similar T cell differentiation into exhausted phenotypes as mentioned here, exhaustion in the TME is a subtly but importantly unique form of exhaustion which will be focused on throughout this review.

The different stages of exhausted CD8⁺ T cells differ greatly based on the expression levels of specific transcription factors and T cell inhibitory receptors. The progenitor exhausted T cells (T_ex ^prog1) have a high proliferative potential with stem-like properties but low cytolytic ability (Jiang et al., 2021). In contrast, terminally exhausted CD8⁺ T cells (T_ex ^term) have a low proliferative potential but high cytotoxicity (Jiang et al., 2021). The terminally exhausted CD8⁺ T cell subset is characterized by higher PD-1, Tim-3 and CD38 expression, and lower CXCR5, CD44, and Ly108 expression than progenitor CD8⁺ T_ex cells (Jiang et al., 2021). Furthermore, terminally exhausted CD8⁺ T cells overexpress the transcription factor TOX and do not express the transcription factors TCF1 and T-bet (Beltra et al., 2020). The differentiation into terminally exhausted CD8+T cells varies as well. For example, progenitor exhausted CD8⁺ T cells do not always turn into terminally exhausted subtype. Some reports suggest the existence of multiple intermediate cell subsets during this progression, if the progression occurs at all (Yang and Wang, 2021; Chen et al., 2022). To further outline the transcriptional heterogeneity of exhausted T cells, a new exhausted CD8⁺ T cell subset was identified recently which expresses unique natural killer (NK) cell related genes (i.e. Klr, Fcg2rb) and surface proteins (i.e. NK1.1, Ly49I, NKG2D) (Giles et al., 2022b).

The distinction between exhausted T cell subtypes is critical because although progenitor exhausted CD8⁺ T cells are responsive to ICBs, terminally exhausted cells are not (Kallies et al., 2020; Jiang et al., 2021). This distinction in part explains why such a large proportion of people fail to reach a complete response to ICB therapy, and why many people eventually progress with such treatments (Kallies et al., 2020; Jiang et al., 2021). One specific reason for this is that when antigen concentration remains high, exhausted CD8⁺ T cells which initially responded to checkpoint inhibitor therapy can become “re-exhausted” (Pauken et al., 2016). Pauken et al. demonstrated this finding in a mouse model of lymphocytic choriomeningitis virus (LCMV) with anti-PD-1 therapy. They found that CD8⁺ T cells failed to acquire memory after immune checkpoint therapy, even if exhausted T cells were reinvigorated (Pauken et al., 2016). This represents an important truism in cancer immunotherapy: once cells reach terminal exhaustion, it cannot be undone and likelihood of ICB resistance increases (Pauken et al., 2016).

Some immune receptors on T cells have been shown to modulate the expression levels of the inhibitory receptors that ICB therapy is trying to target, unveiling potential new approaches to immunotherapy. For example, our group demonstrated that the self-ligand immune receptor SLAMF7 was co-expressed with multiple inhibitory receptors (i.e. PD-1, Tim-3, LAG3) and when SLAMF7 signaling was induced in CD8⁺ T cells, the expression of PD-1, Tim-3, and LAG3 inhibitory receptors were upregulated (O’Connell et al., 2021). Also, SLAMF7 activation resulted in enhanced expression of several T cell exhaustion-promoting transcription factors and epigenetic regulators (O’Connell et al., 2021). We further demonstrated that presence of SLAMF7 on tumor-associated macrophages (TAMs) was necessary for TAMs to induce CD8⁺ T cell exhaustion during their cross-talk with CD8⁺ T cells (O’Connell et al., 2021). Accordingly, SLAMF7 expression on certain TAMs may be a prognostic factor in clear cell renal cell carcinoma (ccRCC) (O’Connell et al., 2021).

It is now widely recognized that exhausted CD8⁺ T cells represent a distinct state in CD8⁺ T cell differentiation and, as such, have a unique profile of transcription factors (TFs) and epigenetic factors that drive and enforce this state via epigenetic remodeling and other mechanisms (Sen et al., 2016; Yao et al., 2019; Ranzoni et al., 2021). Understanding the mechanisms that regulate the expression of specific TFs, their temporal dynamics during the process of T cell exhaustion, and the cellular processes regulated by each TF has critical implications. These details help us not only understand the complex process of exhaustion, but also unveil how to pharmacologically alter this process for clinical benefit. A myriad of TFs have been implicated in regulating T cell exhaustion, some better defined than others, and a complete in-depth review of each can be found elsewhere (Wherry and Kurachi, 2015). There are also a number of TFs suspected to regulate exhaustion which have not yet been formally investigated, including E2F2 (Giles et al., 2022a), and likely others. Our group recently found that the immune cell surface receptor SLAMF7 is capable of regulating some of the critical exhaustion-defining TFs and epigenetic modifiers that regulate different aspects of T cell exhaustion (O’Connell et al., 2021). Below we detail what is currently known about each of these TFs and chromatin modifiers and how they regulate CD8⁺ T cell exhaustion programs.

Recent studies have demonstrated that T cell exhaustion is commonly associated with significant and unique genome-wide epigenetic remodeling programs which differentiate T_ex cells from naïve, effector, and memory T cells (Pauken et al., 2016; Sen et al., 2016; Philip et al., 2017). These epigenetic programs are associated with changes in expression of specific TFs, histone modifications, DNA methylation, and chromatin accessibility. For example, multiple epigenetic profiling studies in tumor infiltrating T_ex cells demonstrated altered chromatin accessibility in thousands of differentially accessible regions, including decreases in chromatin accessibility of genes associated with effector T cell differentiation, such as IFNγ, TNFα, IL-2, KLRG1, and others, and increases in chromatin accessibility of genes associated with T cell exhaustion, such as PD-1, Tim3, LAG3, TIGIT, TOX, Tcf7, NFAT, and others (Sen et al., 2016; Abdel-Hakeem et al., 2021; Giles et al., 2022a; Belk et al., 2022). These changes in chromatin accessibility were also associated with extensive transcriptional alteration in T_ex cells. Importantly, the transcriptional programs and chromatin profiles of T_ex cells were largely conserved between murine T_ex cells of LCMV-infected mice and human T_ex cells derived from cancer and chronically-infected patients (Bengsch et al., 2018; Pritykin et al., 2021), revealing common differentiation programs and regulatory mechanisms. These results also demonstrate that T cell alteration in tumors and chronic infections follows a shared epigenetic differentiation trajectory. In addition to having a similar transcriptional and chromatin profile, studies using murine tumor models identified CD38 and CD101 as differentiating markers for T_ex reprogrammability in PD-1 high CD8⁺ T cells. Specifically, PD-1^HiCD101^LowCD38^Low was a plastic therapeutically reprogrammable chromatin state and PD-1^HighCD101^HighCD38^High was a fixed dysfunctional state of exhausted CD8+T cells (Philip et al., 2017). These findings demonstrate that the epigenetic state of T_ex cells is highly regulated and heterogenous, and this state governs the phenotype and function of T_ex cells, as well as their response to ICB therapy (Gyorki et al., 2009).

It is now well appreciated that the terminal epigenetic state of T_ex ^term is fixed and irreversible, even after reactivation, curative therapy, or following ICB therapies (Pauken et al., 2016; Hensel et al., 2021; Tonnerre et al., 2021; Yates et al., 2021). For example, unlike memory CD8⁺ T cells which robustly expanded upon restimulation, CD8⁺ T_ex ^term cells from anti-PD-1-treated tumor-bearing mice failed to respond to secondary stimuli (Pauken et al., 2016). These results were associated with persistent PD-1 upregulation and a specific DNA methylation signature at the Pdcd1 and other gene loci (Pauken et al., 2016). Also, anti-PD-1 therapy was able to only change 10% of the epigenetic landscape to T_ex ^term cells, implicating that the fixed chromatin state of T_ex ^term cells is a major limitation for anti-PD-1 therapy (Pauken et al., 2016). Similarly, the chromatin accessibility of T_ex ^term cells derived from HIV- and HCV-infected patients remained unchanged before and after resolution of the infection, and in particular at the Tox and Pdcd1 loci, confirming a fixed epigenetic state in T_ex cells after they reach terminal exhaustion (Abdel-Hakeem et al., 2021; Yates et al., 2021). Additionally, and due to the nature of the fixed chromatin state of T_ex ^term cells, when T_ex cells from cured HCV patients were re-challenged with HCV virus, robust upregulation of T cell exhaustion-related genes like PD-1 occurred (Abdel-Hakeem et al., 2021). This phenotype was also associated with reduced expression of genes that regulate effector T cell activity, which further impaired their effector functions. These results necessitate the need to develop novel therapeutic approaches that increase the epigenetic plasticity of T_ex cells. It is important to note that the T_ex cell differentiation state, which is enforced at the epigenetic level, defines specific subsets of T_ex cells. For example, T_ex ^prog1 (Ly108+CD69⁺) and T_ex ^int (Ly108+CD69⁻), which can proliferate and self-renew with ICB therapy (Gyorki et al., 2009) have accessible chromatin in various gene loci that is associated with T cell stemness, such as the Ly108 (SLAMF6) and TCF1 (Tcf7) gene loci (Zehn et al., 2022). In contrast, the terminally exhausted ((Ly108⁻CD69⁺); T_ex ^term) T cell subset, which does not respond to ICB therapy, has accessible chromatin at the gene loci of NR4A, EOMES, and TOX TFs and PD-1, TIM-3, LAG3, CD101, and CD38 T cell inhibitory receptors (Hudson et al., 2019). These data reveal intricacies within the exhausted T cell population, and their implications in cancer progression and tumor evasion of ICB therapy.

As mentioned throughout this review, resistance to ICB and CAR-T cell therapy is a critical clinical problem requiring further research. Scientists have discovered that this resistance, at least in part, occurs due to epigenetic changes in T cell differentiation and in particular in exhausted CD8⁺ T cells. Thereby, it is hypothesized that certain epigenetic-targeting drugs could be applied to T_ex cells as a means of restoring stemness. Indeed, there are various classes of epigenetic modifiers, both in development and FDA approved, which will be discussed here along with their current utility. It is important to note that most of these epigenetic modifiers do so on a global scale, so specificity to a subset of immune cells in the tumor microenvironment is currently limited. Figure 3 demonstrates how these therapies can be conceptualized in addition to the altered epigenetic modifiers discussed here.

Broad-spectrum epigenetic inhibitors target genome-wide cancer specific gene expression and include inhibition of various targets. Specifically, inhibition of DNA methyltransferase (DNMT) (Chiappinelli et al., 2015; Giri and Aittokallio, 2019), histone deacetylase (HDAC) (Eckschlager et al., 2017; Lue et al., 2019), and bromodomain and extra-terminal motif proteins (BET) (Kong et al., 2021; Shorstova et al., 2021) are the most common. DNMT inhibitors target hypermethylated DNA regions, and thus transcriptionally inactive chromatin. This reversal of silenced gene transcription by loss of DNA methylation allows the genes to become active again, which is desirable in the setting of tumor suppressor gene transcriptional silencing. DNMT inhibitors have been used in treating blood malignancies for years (Gore, 2005). Specifically, decitabine and azacytidine, two DNMTi drugs, are approved for treating myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML). The drugs’ activity towards malignant cells supports the importance of epigenetic regulation in tumors. More recently, DNMTi use has been found to affect T cell exhaustion (Wang et al., 2021). Wang et al. found that that decitabine epigenetically reprogrammed CAR-T cells, enhancing and prolongating their antitumor activity (Wang et al., 2021). Furthermore, other groups showed that DNMTi therapy reversed exhausted CD8⁺ T cells and enhanced their effector function in different cancer models (Chiappinelli et al., 2015; Roulois et al., 2015). DNMT inhibitor therapy with various different drugs are currently being tested in clinical trials for multiple hematologic and solid malignancies, most in combination with another agent such as chemotherapy, immunotherapy, or second epigenetic modifying drug (Gore, 2005; Hu et al., 2021; Baer et al., 2022).

HDAC inhibitors on the other hand, preserve acetyl groups on lysine residues in histone tails, allowing open chromatin configurations and thus favoring gene transcription (Laino et al., 2019). Indeed, HDAC inhibitors cause apoptosis and arrest tumor cell growth and potential angiogenesis. There are multiple HDAC isoforms, and some inhibitors target specific HDAC isoforms while others are global inhibitors (Laino et al., 2019). For example, selective HDAC6 inhibition with ACY-1215 and ACY-241 was efficacious towards melanoma patients’ T cells ex-vivo, harvested from either peripheral blood or tumor biopsies (Laino et al., 2019). Specifically, HDAC6 inhibitor increased T cell effector functions, reduced exhaustive markers, and in turn enhanced the killing of malignant cells (Laino et al., 2019). Importantly, higher frequencies of T effector memory cells remained after treatment with ACY-1215 as well (Laino et al., 2019). Chromatin accessibility was increased in genetic regions associated with such effector and memory T cell functions. Furthermore, McCaw et al. found that an HDAC inhibitor, specifically the HDAC1 inhibitor etinostat (ENT), promoted immune-modulatory molecule expression which in turn re-sensitized tumors to ICB (McCaw et al., 2019). These findings support that epigenetic targeting can reprogram exhausted T cells and enhance effector function of T cells once resistant to ICB.

Another global inhibitor of epigenetic processes are bromodomain and extra-terminal domain (BET) inhibitors. BET overexpression can contribute to carcinogenesis, such as in hyper-acetylation of proliferation-promoting genes or hyper-acetylation of oncogene enhances (Jin et al., 2021). BET inhibitors recognize and bind to acetylated lysine residues, altering transcriptional activation and chromatin remodeling (Shorstova et al., 2021). Indeed, mouse studies revealed that suppression of BET proteins reduced AML burden (Zhong et al., 2022). Moreso, the specific BET inhibitor JQ1 reversed T cell exhaustion from ICB therapy in AML T cells in vitro and in vivo (Zhong et al., 2022). This BET inhibitor also reduced PD-1 and Tim-3 expression and increased cytokine secretion by AML patient-derived T cells (Zhong et al., 2022). As exciting as this data is, whether BET inhibition can reverse exhaustion in CML in vivo models which are resistant to anti-PD-1 therapy remains to be seen, and would be an important next step. Using a chronic lymphocytic leukemia (CLL) model, Kong et al. did demonstrate that BET inhibition reversed CAR-T cell exhaustion as measured by multiple factors including reduced inhibitory receptor expression and increasing their proliferation abilities (Kong et al., 2021).

Other methods of epigenetic modulation include narrow spectrum inhibitors. One of these methods focuses on EZH2 inhibition, due to its methyltransferase activity. Wang et al. found that EZH2 depletion reduced MYC expression and hindered neuroblastoma and small cell lung carcinoma tumor growth (Wang et al., 2022). However, they found that this EZH2 inhibition was independent of its methyltransferase activity (Wang et al., 2022), revealing that EZH2 may play a role in cancer beyond its enzymatic epigenetic reprogramming abilities. In 2020, the first EZH2 inhibitor was FDA approved for patients with relapsed or refractory follicular lymphoma. To receive this treatment, patients require a confirmed EZH2 mutation, as the studies determined a higher response rate in patients with confirmed EZH2 mutation in comparison to WT (69% vs 35%) (Morschhauser et al., 2020). However, the therapy was found to be tolerable and produce reasonably efficacy independent of EZH2 mutation (Morschhauser et al., 2020). Thus, it could be argued that this could be an option for refractory lymphoma patients (or other B cell malignancies) without EZH2 mutation, given its reasonable efficacy and safety profile.

Different EZH2 inhibitors exist, most targeting the methyltransferase abilities of EZH2. For this reason, it has been proposed as a promising combination with cytotoxic chemotherapy, as it makes DNA more accessible for DNA-damaging agents (Adema and Colla, 2022). In theory, one could use a lower dose of cytotoxic drug with a largely not toxic EZH2 inhibitor. This would improve safety profiles of chemotherapy, especially in older individuals (Adema and Colla, 2022). Furthermore, it can be combined with other epigenetic modifiers for better results. For example, the combination of EZH2 inhibitor with HDAC inhibitor increased lymphoma cell line apoptosis (Lue et al., 2019). The impact of combination epigenetic reprogramming on CD8⁺ T cell exhaustion and possible synergistic effects in vivo are unclear and warrant further investigations.

Outside of targeting epigenetic enzymes, research has supported targeting important exhaustion-driving transcription factors, and preclinical results have been promising. For example, CAR-T cells with Nrf4A family proteins knocked out had improved tumor-directed efficacy in mouse models, and knockout of TOX improved CAR-T cell efficacy as well (Tang et al., 2021). Although there are no current TOX inhibitors, this would seem to be a viable option for reversing ICB resistance. Another option could be to use an upstream regulator of TOX. Specifically, a regulator such as SLAMF7 which regulates TOX expression as discussed above. Thus, we propose that epigenetic directed therapy in combination with a SLAMF7 inhibitor may be a way to reduce CD8+T cell terminal exhaustion and warrants future investigations.

It is no question that the discovery of checkpoint inhibitors and development of CAR-T cell-based therapy have drastically changed the way clinicians approach cancer treatment and management. However, despite great success, many tumors unfortunately remain insensitive to such therapies, with solid tumors remaining particularly challenging to treat via immune-based mechanisms. Recently, the discovery of epigenetic changes driving T cell differentiation toward exhaustion has unveiled that epigenetic factors reinforce CD8⁺ T cell exhaustion as well. These studies, together with known transcription factors driving CD8⁺ T cell exhaustion, support that the exhausted CD8⁺ T cell phenotype is at least in part due to epigenetic modulations. Thus, the utilization of epigenetic modifying drugs either as single agents or in combination with potent therapies might improve the effector and memory responses of CD8⁺ T cells.

As important as the understanding of epigenetic remodeling of T_ex cells is for the tumor immunology field, how these drugs will translate in vivo remains to be seen. For example, many of these drugs are global modifiers of specific epigenetic processes. Thus, they would need to be directed towards exhausted CD8⁺ T cells specifically to confirm these theories. Utilization of a drug-antibody conjugate for an inhibitory receptor highly expressed on exhausted CD8⁺ T cells, such as PD-1, may be one way to localize epigenetic modifying therapies to the T_ex immune cell subset in the TME. In practice, it may be more beneficial to epigenetically modify both cancer cells and T_ex cells, in which case a purely T cell-targeted approach would not be as desirable, but this remains to be seen. Not only may the inhibitors have effects on tumor cells, but they also can have off target effects on healthy cells and tissues unrelated to cancer. In addition, it is well-known that potent immune activation in the setting of cancer can trigger autoimmune-like reactions (Young et al., 2018). Thus, ensuring the therapy is selective for exhausted CD8⁺ T cells is an important limitation of current studies and should be further addressed.

This work leads one to wonder whether the combination of broad-spectrum epigenetic regulators in combination with ICB or CAR-T cells therapy would be optimal in cancer patients who do not respond to ICB alone, and some studies already support this hypothesis (Eckschlager et al., 2017; Li et al., 2018; McCaw et al., 2019; Wang et al., 2021; Zhong et al., 2022). Whether it is due to reversing CD8⁺ T cell exhaustion or some other factor remains to be seen. The lack of specificity with global epigenetic regulators may limit their efficacy and makes mechanistic understanding of the inhibitors in relation to exhaustion difficult to discern. This muddies identifying patients who will respond to the therapy versus those who will not. Currently, genetic testing of specific epigenetic regulators is being done on patients to predict their response to such inhibition. However, as we have learned from targeted therapy approaches, because a gene is altered does not mean that targeted inhibition will reverse the phenotype. Indeed, it is unclear whether patients without such mutations could also respond to therapy, which would support that these epigenetic therapies could be efficacious for a larger population. Given patients with reduced options, we propose that epigenetic approaches should be considered. The authors here believe that this is a field of study with great promise and may be a possible solution to ICB or CAR-T cell resistance. Further research is necessary to unweave the intricate epigenetic mechanisms in exhausted CD8⁺ T cells, improve specificity of epigenetic modifying therapy, and discover optimal approaches for reversing terminal CD8⁺ T cell exhaustion in the TME.