Strategies to improve CAR-T cell dysfunction

Over the past decade, we have made significant progress in improving the efficacy of CAR-T cell therapies. However, its clinical benefit remains limited, especially in solid tumors. Even among hematological malignancies, patients who respond to CAR-T therapy are still at risk of relapse due to multiple factors. This is due to the frequent dysfunction of T cells, especially CAR-T cells, in cancer, and many evasion mechanisms impair anti-tumor immunity.

Compared with dysfunctional T cells, functional T cells exhibit a series of activities under antigen stimulation, including (1) expansion; (2) secretion of effector cytokines (such as IL-2, IFN-γ, TNF-α, perforin and granzyme) and lyse target cells; (3) still survive after removal of antigen stimulation; (4) still have the above response during secondary antigen stimulation. It can be considered that if at least one of the above criteria is not met, the T cell is considered dysfunctional.

Currently, the most studied T cell dysfunctional state is exhaustion, which is characterized by the loss of all the above-mentioned properties of functional T cells. Another type of dysfunction is T cell senescence, which occurs when T cells permanently halt their cell cycle and proliferation while maintaining their cytotoxic capabilities.

T cell dysfunction is a major obstacle to the development of CAR-T therapies, which is a problem encountered from the beginning of the cell manufacturing step. Including the design, in vitro operation and culture conditions of CAR, all may play a key role in this. Therefore, engineering CAR-T cells to improve their survival and reverse or prevent the exhaustion phenotype may improve CAR-T cell therapy. To date, some research has taken some approaches to confining and creating exhaustion-resistant T cells, and they are being tested in several pivotal clinical trials. These methods may improve the dysfunction of CAR-T cells and bring hope to patients.

T cell exhaustion

T cell exhaustion is closely related to the continued driving effect of antigen. Under conditions of acute infection, naive CD8+ T cells differentiate into short-lived effector cells (SLEC) or memory precursor effector cells (MPEC). After antigen clearance, most SLECs die, while MPECs survive to form memory CD8+ T cells, forming long-term protective immunity.

Under chronic infection, it was observed that high-affinity antigen-specific CD8+ T cells differentiated into SLEC and were rapidly cleared and died, but no MPEC subpopulation was formed. In contrast, CD8+ T cells targeting low antigen affinity expand, become exhausted, manifested by significant reductions in proliferation, cytotoxicity, and cytokine production, and die in stalemate against persistent antigens.

A common feature of chronic viral infections and cancer is that both are long-term diseases characterized by the persistent effects of antigens. Naïve CD8+ T cells targeting tumor antigens are first initiated in peripheral lymphoid tissues to generate stem cell-like PD-1loCD8+ T cells with self-renewal properties. Under the action of chemokines CCL5 and CXCL9, they migrate to the TME and form effector PD-1loCD8+Tex. However, continued antigen load in the TME ultimately forces these cells to continue to differentiate into nonfunctional PD-1hiCD8+Tex. The PD-1hi state is accompanied by increased expression of co-inhibitory receptors (including Tim-3, LAG-3, CD160, 2B4, TIGIT, and CTLA-4) and a progressive loss of effector function. Once CD8+Tex enters the PD-1hi state, epigenetic enforcement blocks dedifferentiation back to a functional stem cell-like and effector-like PD-1lo state. The anti-tumor response promoted by ICB (e.g., anti-PD-1) comes only from the expansion of the PD-1loCD8+Tex subpopulation in the lymphatic or tumor, and the low-function, ICB-resistant PD-1hiCD8+Tex will eventually undergo apoptosis.

Both tumor- and chronic virus-specific CD8+ T cells had significant enrichment of genes related to TCR signaling (Batf, Egr2, Ezh2, Irf4, Nfatc1, Nfatc2, Nr4a1, Nr4a2, and Nr4a3). This observation further supports the persistence involvement of antigen as a major driver of depletion.

T cell senescence

Cell senescence is a typical multi-cause process that occurs in a variety of cell types and is characterized by cell cycle arrest. T cell senescence and T cell exhaustion are two distinct processes. Senescent cells exhibit a specific senescence-associated secretory phenotype (SASP) that exerts paracrine effects on other cells. Aged T cells often exhibit a CD45RA+CD27−CD28−KLRG1+CD57+ phenotype. They lose the ability to proliferate and secrete IL-2, but express cytolytic molecules, IFN-− and TNF-α, and have Cytotoxicity.

T cell senescence is often associated with activation of the DNA damage response (DDR), which is triggered by telomere erosion, DNA damage caused by reactive oxygen species (ROS), glucose or growth factor deprivation, and activation of the cAMP pathway. Senescent T cells exhibit increased p38 activation and reactive oxygen species accumulation, impaired mitochondrial activity, downregulation of mTOR signaling, and telomere shortening. Despite undergoing cell cycle arrest, these cells remain cytotoxic, similar to terminal effector T cells.

In addition, repeated activation during chronic infections or autoimmune diseases has also been associated with T cell senescence. Tumor-induced senescent T cells have been shown to suppress the activity of other T cells in the tumor microenvironment. Increasing evidence suggests that tumor-induced T cell senescence is an evasion mechanism.

Mechanisms of CAR-T cell dysfunction

Transduced CAR-T cells exhibit similar dysfunction mechanisms as unmodified T cells. In fact, CD8+ CAR-T cells and TILs isolated from the same tumor-bearing mice have similar transcriptional and epigenetic profiles. Targeting pathways known to contribute to T cell dysfunction may improve CAR T cell function.

Design of CAR

The type of extracellular CAR domain, the type of spacer connecting the single-chain antibody to the transmembrane domain, and the co-stimulatory domains that make up the CAR have been shown to be important. T cell function and persistence have profound effects. Most CARs used in clinical trials are derived from mice, which may result in humoral and CD8+ T cell-mediated immune responses leading to immune rejection.

The affinity of CAR affects tonic signaling, and excessive affinity to target cells may lead to T cell exhaustion. One study developed a CAT-CAR (CD19 single-chain antibody) with a 40-fold lower affinity for CD19 than conventional CARs. The construct was tested in pediatric patients with relapsed or refractory BLL, and clinical studies found that 11 of 14 patients with increased CAR-T cell expansion showed persistence.

In addition, the costimulatory domain also plays a key role in the persistence and effectiveness of CAR-T cells. Costimulatory molecules include CD28, ICOS, CD27, 4-1BB, OX40 and CD40L. Several studies found enhanced T cell persistence in CAR constructs equipped with the 4-1BB domain instead of the CD28 domain. Other studies reported that 4-1BB-containing CAR T cells displayed an enhanced memory T cell (TEM) phenotype, which may delay CAR T cell exhaustion. ICOS belongs to the CD28 costimulatory molecule family. Studies have proven that the combination of ICOS and 4-1BB costimulatory domains in the CAR structure significantly increases the persistence of T cells. These observations suggest that the structure of the costimulatory domain is an important factor influencing CAR-T cell persistence.

In vitro operations

The process of making T cells requires obtaining a sufficient number of healthy T cells from the patient. However, it is possible to obtain senescent T cells from older patients or patients pre-treated with chemotherapy/total irradiation. Especially those chemotherapy containing clofarabine or doxorubicin, resulting in lymphopenia, will make the quality of the final CAR-T cells unsatisfactory, and the type of T cells used for infusion will seriously affect the effect of the treatment.

In fact, better persistence was obtained in preclinical models by normalizing the CD4/CD8 ratio or using naive or memory cell subsets. Several studies have shown that using early-lineage cell-enriched T cell populations enables better expansion and improves the durability and efficacy of CAR-T cell therapy. Additionally, multiple studies have shown that adding antioxidants, such as N-acetylcysteine, to cell cultures during the manufacturing process can also inhibit effector differentiation and promote the expansion of TSCM cells.

Immunosuppressive microenvironment

In the TME, the presence of immunosuppressive cytokines produced by myeloid-derived suppressor cells (MDSCs), cancer-associated fibroblasts (CAFs), and tumor cells Can lead to T cell exhaustion.

Chronic exposure to antigens can also lead to T cell exhaustion. If antigen stimulation persists, T cells will undergo a series of epigenetic, metabolic and transcriptional changes, indicating an exhaustion state. This process occurs in a progressive manner, with IL-2 and TNF-α being lost early and IFN-γ and chemokine secretion decreasing later in the progression of depletion. Although high proliferative capacity is also lost early, exhausted T cells can still proliferate to a limited extent when stimulated in vivo. Depleted cells also showed high expression of inhibitory receptors such as PD-1, TIM-3, LAG-3, CD160, BTLA, CTLA-4 and TIGIT.

Strategies to improve CAR-T cell dysfunction

Expression of cytokines and their receptors

Fourth generation CAR T cells have recently been developed for Resist the immunosuppressive environment in the TME while overcoming immune depletion. The TRUMKS design combines the cytotoxic functions of CAR-T cells with the in situ delivery of cytokines with immunomodulatory capabilities. Under the action of the induction system, after the CAR binds to the antigen, cytokines are synthesized and act in an autocrine manner to increase the survival and expansion of T cells. Cytokines can also act in a paracrine manner, modulating the surrounding environment and interfering with immunosuppressive cytokines present in the TME. A series of cytokines including IL-12, IL-7, IL-15, IL-18, IL-21 and IL-23 are currently under research and have entered early clinical trials.

IL-12 is a pro-inflammatory cytokine that can induce Th1CD4+ T cell responses and promote the expansion and persistence of CD8+ clones. It is also responsible for regulating the cytotoxic activity of CTL and natural killer (NK) cells, reactivating anergic tumor-infiltrating lymphocytes, recruiting NK cells, and suppressing Tregs. Preclinical studies of CD19-CAR-T cells constitutively expressing IL-12 demonstrated enhanced tumor killing efficacy and immune memory against cancer antigens. However, the potentially lethal toxicity associated with IL-12 necessitates the development of an inducible system that limits IL-12 secretion only upon CAR activation. Several clinical studies (NCT02498912, NCT03932565, and NCT03542799) are ongoing and recruiting.

IL-15 is a cytokine that stimulates the activation, proliferation and cytotoxic activity of CD8+ T cells and NK cells. IL-18 increases Th1 cell cytokine production while inhibiting IL-10 synthesis. IL-15 and IL-18 are both immune response enhancers and have been tested on CAR-T cells. Compared with conventional CAR-T cells, IL-18- and IL-15-secreting CAR-T cells showed enhanced expansion and persistence in tumor-bearing mice and enhanced tumor cell proliferation in vitro and in vivo toxicity. Multiple clinical trials are currently recruiting to test the role of IL-15- and IL-18-secreting CARs in solid and hematological tumors using engineered T cells and NK cells.

Recent studies have shown that it is possible to co-express multiple cytokines in the same CAR-T cell. NCT04833504 is a recently completed clinical trial in which CD19+ CART cells expressing IL-7 and CCL19 were tested in patients with relapsed or refractory B-cell lymphoma, but the results have not yet been reported. Two additional clinical trials are currently recruiting to test IL-7-expressing CAR T cells in combination with secretion of PD-1 blockade or other cytokines.

Combination checkpoint blockade therapy

Strategies to reduce exhaustion by inhibiting checkpoint signaling include knocking out co-suppressor molecules via shRNA expression vectors or CRISPR/Cas9. Many studies report that blocking checkpoint inhibition restores cytokine production and promotes CAR-T cell survival. In addition, simultaneous blockade of multiple immune checkpoints, such as PD-1, TIM-3, and LAG-3, can synergistically increase the effector function of CAR-T cells.

Combining CAR-T cells with immune checkpoint blockade therapy may be an effective strategy to enhance anti-tumor activity, persistence and memory cell formation. Anti-PD-1 antibodies enhanced the anti-tumor activity of anti-HER2 CAR-T cells in glioblastoma and breast cancer cell lines. However, some clinical trials have reported negative results after using a combination of PD-1 inhibitors and anti-GD2 CAR-T cells to treat neuroblastoma patients.

In addition, some studies have adopted other strategies to block PD-1, such as gene editing to make CAR-T cells secrete PD-1 blocking antibodies or downregulate PD-1.

Using Stem T Cells

Effector T cells were initially considered ACT's best product because of their ability to kill tumor cells. However, they have limited persistence, poor expansion capacity, and are susceptible to depletion.

TSCM is an ideal candidate for ACT due to its long lifespan, strong self-renewal capacity, and ability to differentiate among different T cell populations. Clinical studies have shown that infusion of phenotypic and functional TSCM-like CAR-T cells (CD62L+, CD28+, and CD27+) can produce favorable results. For example, CLL and multiple myeloma patients treated with anti-CD-19 CAR-T cells showed favorable responses associated with the CD27+CD45RO−CD8 cell population. Furthermore, studies in mice show that infusion of CD62L+-enriched T cell populations increases expansion and persistence, leading to durable tumor regression.

Summary

Although CAR-T cell therapy has made significant progress in the past decade, the possible dysfunction of CAR-T cells in patients results in limited persistence. remains a challenge, mainly due to T cell exhaustion and senescence. By designing the CAR structure, changing production conditions, or introducing new treatments, we may create engineered T cells that are resistant to exhaustion, thereby further expanding the clinical application of CAR-T cells. While results to date remain limited, the possibilities are endless and a breakthrough may be on the horizon.