Trogocytosis in CAR immune cell therapy: a key mechanism of tumor immune escape

Abstract

Immune cell therapy based on chimeric antigen receptor (CAR) technology platform has been greatly developed. The types of CAR immune cell therapy have expanded from T cells to innate immune cells such as NK cells and macrophages, and the diseases treated have expanded from hematological malignancies to non-tumor fields such as infectious diseases and autoimmune diseases. Among them, CAR-T and CAR-NK therapy have observed examples of rapid remission in approved clinical trials, but the efficacy is unstable and plagued by tumor resistance. Trogocytosis is a special phenomenon of intercellular molecular transfer that is common in the immune system and is achieved by recipient cells through acquisition and internalization of donor cell-derived molecules and mediates immune effects. Recently, a novel short-term drug resistance mechanism based on trogocytosis has been proposed, and the bidirectional molecular exchange between CAR immune cells and tumor cells triggered by trogocytosis partially explains the long-term relapse phenomenon after treatment with CAR immune cells. In this review, we summarize the research progress of trogocytosis in CAR immunotherapy, discuss the influencing factors of trogocytosis and its direct and indirect interference with CAR immune cells and emphasize that the interference of trogocytosis can further release the potential of CAR immune cell therapy.

Introduction

In the past century, radiotherapy, chemotherapy, surgical treatment and other methods are the main methods of tumor treatment. These methods usually focus on the lesion and achieve the purpose of clearing the lesion or reducing the volume of the tumor by killing the tissues or cells at the specified location [1]. For advanced or refractory patients, the treatment is usually palliative, with the purpose of reducing the suffering of patients, prolonging the survival time and improving the quality of life [2, 3]. Nevertheless, with the research and development of tumor molecular science, the importance of the molecular characteristics of tumor and immune cells have been highlighted, and the traditional treatment concept centered on the physical location of the tumor has been challenged [4–6].

Under the guidance of precision medicine and precision oncology, the continuous development of immunotherapy has brought hope to patients with malignant tumors. The development and application of immune checkpoint inhibitors [7], cancer vaccines [8], immune cell adoptive therapy [9] and other therapies have benefited many patients, among which the chimeric antigen receptors (CARs)-based immune cell therapy platform is booming [10, 11]. At the heart of this platform is a synthetic chimeric antigen receptor, which is characterized by the direct activation and relocalization of controlled immune cell activity [12]. Through the combination of CAR molecules and pathogenic cell surface antigens, signals are transmitted to the intracellular signal domain of CAR, which directly activates immune cells, such as T cells, NK cells, macrophages, Treg cells, neutrophils, etc., to produce corresponding activities or functions to clear target cells and regulate the immune microenvironment [13]. CAR-T cell therapy is the most mature technology in the CAR immunotherapy platform, and multiple products have been launched for the treatment of hematological malignancies, and the approved indications include leukemia, lymphoma, and myeloma [14, 15].

Currently, CAR-T therapy is highly personalized, and patient autologous cells are the only available clinical cell source, which represents a complex manufacturing process and high treatment cost [16]. CAR-NK therapy has been proposed subsequently because NK cells do not have strict requirements for HLA matching and can be applied to a wide range of patients, promising to be developed as a universal cell therapy product, and the limited persistence capacity and cytokine profile imply minimal side effects [17]. In addition, CAR-T is not effective in the treatment of solid tumors, the dense physical barrier hinders CAR-T infiltration. Even if the intratumoral delivery of CAR-T cells is achieved by technical methods, the immunosuppressive tumor microenvironment and continuous antigenic stimulation induce CAR-T cell exhaustion [18]. The study of CAR-NK shows the potential of innate immune cells in the application of CAR platform [19]. Macrophages are a bridge connecting innate and adaptive immunity, and have tumor homing effect, phagocytosis, cytokine secretion and antigen presentation functions [20]. The construction of CAR-Macrophage (CAR-M) enables these functions to form a network of targeted inhibition in solid tumors, which has made progress in clinical studies and is a strong candidate for the treatment of solid tumors [21, 22]. In addition, other CAR immune cells, such as CAR-Treg and CAR-neutrophils, are also in preclinical research [23, 24], and the CAR immune cell technology platform has been promoted for the treatment of autoimmune diseases, viral and bacterial infections, fibrosis, cardiovascular diseases and other non-tumor diseases, and has shown to be safer and more compliant than traditional drugs [25–27]. This means the advent of the era of immune cell therapy.

However, clinical studies and continued follow-up of patients treated with CAR-T products have revealed a significant challenge of CAR-T therapy: tumor recurrence [28, 29]. Now, several studies have confirmed that in the face of CAR-T immune pressure, tumor cells are able to conduct targeted immune escape through multiple mechanisms [30, 31]. Although CAR-T cells can produce a strong therapeutic effect in the early stages of treatment, the long-term results are heterogeneous, with the main negative factors being the exhaustion of engineered T cells and stimulation of cytokine signaling pathways, while the memory-like T cell subsets is still functioning to some extent, but in a unfavorable niche, and the effect is limited. Additionally, tumors can make CAR-T cells lose their targeting by down-regulating antigen expression, and antigen loss relapse occurs [32, 33]. Further research on the mechanisms of tumor immune escape will contribute to the development of novel CAR immune cell therapy products and the optimization of the clinical efficacy of existing products.

The word trogocytosis is a combination of the Greek words “trogo” (meaning “gnaw”) and “cytosis”, which represents the characteristic of trogocytosis, that is, recipient cells quickly “bite” donor cells to obtain ectopic expression of donor cell surface molecules [34]. This phenomenon will lead to changes in the function of the recipient or donor cells and subsequently mediate a variety of physiological effects including antigen presentation, tumor immune escape, intercellular communication, and clearance of pathogenic microorganisms, etc [35, 36]. Now, several studies have provided credible evidence that trogocytosis is an important factor limiting the immune efficacy of CAR cells. Interestingly, the trogocytosis between CAR-T cells and tumor cells is a bidirectional effect. When tumor antigens are transferred to CAR-T cells, CAR molecules are also acquired by tumor cells.

Now, a novel mechanism of bidirectional trogocytosis resistance between CAR immune cells and tumor cells has been proposed, which partially explains the relapse phenomenon after treatment with CAR immune cells. This review summarizes the current advances of trogocytosis in CAR immune cell therapy platforms, discusses the factors influencing its occurrence and its impact on CAR immune cell therapy, and summarizes several feasible approaches to enhance the role of CAR immune cells by blocking trogocytosis. In summary, we highlight the critical role of trogocytosis in immune escape of low antigen-expressing tumor cells and the potential for enhanced clinical therapeutic efficacy brought about by interference with trogocytosis.

Trogocytosis in CAR immune cell therapy

Trogocytosis is a phenomenon of molecular transfer on the surface of intercellular membranes, which is widespread in the human body and can occur between multiple cell types [37, 38]. In essence, trogocytosis is also a kind of phagocytosis, but it is characterized by “nibbling” on part of the cell membrane of the donor cell, rather than whole-cell phagocytosis [39, 40]. The correspondence is that the donor cells lose these membrane proteins and their corresponding functions, while the recipient cells gain the membrane proteins and produce the corresponding alternative activities [41, 42]. In some cases, some intracellular substances can also be transferred between cells by the trogocytosis pathway but have not been studied in more depth [43]. Trogocytosis was first discovered in Amoeba and has subsequently been observed in the immune system, for example, MHC molecules are transferred from antigen-presenting cells to T cells via trogocytosis, and are involved in T cell activation and various T cell-mediated diseases [44]. In addition, trogocytosis is a non-APC pathway for immune cells to acquire antigens, which allows immune cells to acquire antigen information quickly in neighboring cells and enhance immune surveillance. For example, peripheral blood NK cells of cancer patients acquire tumor antigens through trogocytosis and produce enhanced anti-tumor effects [45]. It is precisely because of the universality of trogocytosis and its importance in the body’s immune activities that tumor cells can cleverly take advantage of this effect to transfer antigens to the surface of immune cells, inhibit the activation of immune cells or mediate the internecine killing of immune cells to form immune escape [45, 46].

Trogocytosis mediated tumor immune escape has been characterized in studies of CAR-T and CAR-NK therapy (Table 1) and has been found to take two forms (Fig. 1). On the one hand, CAR immune cell transfer of tumor antigen is involved. For example, in the studies related to CAR-T and CAR-NK, it has been found that leukemia cells can transfer target antigens to the surface of CAR immune cells through trogocytosis, reduce the expression of antigen on the surface of tumor cells and induce the emergence of target antigen-positive CAR immune cells [47, 48]. On the other hand, tumor cells can acquire CAR molecules via trogocytosis, which leads to antigen masking and loss of CAR molecules from engineered immune cells [49]. In brief, trogocytosis during CAR-T/NK therapy impairs the activity of CAR immune cells and provides a window of opportunity for tumor immune escape. Remarkably, this process is rapid, transient, and reversible [49, 50]. Now, a clinical study of trogocytosis as a predictive marker for CAR-T cell response in diffuse large B cell lymphoma has been approved (NCT06352242) to assess the association of trogocytosis with treatment response or severe side effects.

Table 1.

Summary of preclinical studies on CAR-T/NK cells interfered by trogocytosis

CAR-T/NK	Cancer type	Trogocytic molecules	Direction	effect	Reference
CD19 CAR-T	Leukaemia	CD19	Cancer cell to CAR-T	Trogocytosis mediates tumor cell CD19 antigen loss and ectopic expression of CD19 on CAR-T cells, ultimately leading to CAR-T cell fratricide and exhaustion.	Mohamad Hamieh [43]
CD19 CAR-T	B cell malignancies	CD19	Cancer cell to CAR-T	Trogocytosis is associated with the affinity of CAR molecules, and low-affinity CAR is beneficial for alleviating the effects of trogocytosis.	Michael L. Olson [67]
CD22 CAR-T	ALL	CD22	Cancer cell to CAR-T	Trogocytosis-mediated antigen transfer occurs rapidly and promotes CAR-T cell fratricide.	Xiaoyu Zhou [73]
MSLN CAR-T	Ovarian cancer	MSLN	Cancer cell to CAR-T	Trogocytosis occurs rapidly and correlates with antigen expression, promoting CAR-T cell fratricide and tumor antigen heterogeneity.	Esther Schoutrop [46]
CEA CAR-NK	Colon cancer	CEA	Cancer cell to CAR-NK	Trogocytosis mediates CEA antigen loss, but there is no difference compared to the control group, which is an intrinsic mechanism.	Alexander Sebastian Franzén [54]
CD19 CAR-NK	Lymphoma	CD19	Cancer cell to CAR-NK	Trogocytosis can occur in CAR-NK therapy and mediate tumor antigen loss, CAR-NK cell fratricide and dysfunction.	Ye Li [44]
EGFRvIII CAR-T/ B7H3 CAR-T	Glioma	EGFRvIII CAR/ B7H3 CAR	CAR-T to cancer cell	Trogocytosis mediates CAR molecule loss and subsequent CAR-t cell dysfunction, whereas ectopic expression of CAR molecules on the surface of tumor cells triggers tumor cell antigen masking and promotes tumor resistance.	You Zhai [45]

Abbreviations ALL, Acute lymphocyte leukemia; CAR, chimeric antigen receptor; CEA, Carcinoembryonic antigen-related cell adhesion molecule 5; EGFR, epidermal growth factor receptor; MSLN, Mesothelin

Fig. 1.

Bidirectional trogocytosis effect in CAR-T/NK therapy. In the adoptive therapy of CAR-T/NK cells, after the formation of immune synapse between CAR-T/NK cells and tumor cells, CAR-T/NK cells can obtain tumor cell antigen through trogocytosis, and tumor cells can also obtain the expression of CAR molecules through trogocytosis. This bidirectional trogocytosis effect is rapid, transient and reversible, which can mediate the short-term drug resistance of CAR-T/NK cells and tumor immune escape

CAR-M is a novel treatment method for the solid tumors based on the CAR technology platform, which aims to combat solid tumors by activating the phagocytic activity of macrophages and its positive effects on adaptive immunity, such as promoting antigen presentation, improving T cell infiltration and activation [51–53]. As a typical phagocyte, macrophages can nibble abnormal cells in the human body through whole cell phagocytosis or trogocytosis [54]. In CAR-M, trogocytosis is also found. In 2018, Meghan A Morrissey et al. [55] tested the effect of different intracellular domains on the phagocytic activity of CAR-M and found that trogocytosis was the main phagocytic mechanism of FcRγ, Megf10 and other intracellular domains activated CAR-M rather than the expected whole-cell phagocytosis. Whole-cell phagocytosis of CAR-M is facilitated through tandem intracellular domains (FcRγ-PI3K^P85), but the positive role of trogocytosis in CAR-M is undeniable, as CAR-M activated by a single intracellular activation domain (e.g. FcRγ, CD3ζ) has demonstrated powerful anti-tumor activity [55, 56]. However, whether trogocytosis can mediate membrane transfer of CAR molecules and antigens in CAR-M has not been reported. Given the special identity of macrophage phagocytes and the fact that trogocytosis is one of the main killing modes of CAR-M, the trogocytosis effect of CAR-M is unlikely to interfere with its activity. Nonetheless, it is possible that tumor cells acquire CAR molecules on the surface of CAR-M, and a similar example arises in hematopoietic stem cells, which can acquire macrophage surface proteins through trogocytosis, increasing their retention in the bone marrow [57].

Discovery of trogocytosis mechanism and influencing factors in CAR immune cell therapy

The mechanism of tumor antigen transfer to CAR immune cells via trogocytosis is not well elucidated, but is thought to be related to the synaptic formation, antigen density, and affinity of CAR molecules with target antigens, independent of other CAR structures, downstream signaling, and tumor type (Fig. 2).

Fig. 2.

Influencing factors and regulatory mechanisms of trogocytosis. Trogocytosis is affected by immune synapse formation, CAR molecular affinity and cholesterol metabolism, but the mechanism is not clear. Recent studies have found that tumor-derived factors can promote the expression of ATF3, then down-regulate CH25H, and promote the occurrence of trogocytosis. However, in the case of CAR-T/NK cell fratricide secondary to trogocytosis, the surviving CAR-T/NK^TROG can promote exhaustion phenotype by inhibiting IRF1 and STAT1, and down-regulate HLA expression to escape T cell recognition

Synaptic formation

An interesting phenomenon is that trogocytosis is not related to some specified antigens but is related to the target of CAR. CD19 CAR is responsible for the transfer of CD19 from tumor surface to CAR-T cell surface, while MSLN CAR is responsible for mediating the transfer of MSLN [47, 50]. After suppression of immune synapse formation with actin inhibitors, tumor antigen transfer also disappeared, suggesting that immunosynapse formation between CAR molecules and target antigens is necessary for trogocyrosis [58]. A similar phenomenon has been observed in the study of CD19 CAR-NK [48]. However, this does not mean that the primary trogocytosis effect of NK cells has disappeared. A study on the evaluation of the anti-tumor effect of CEA CAR-NK-92 also found trogocytosis effect, but it is unrelated to the expression of CAR molecules and does not mediate immunosuppression against CEA CAR-NK-92 [59]. Moreover, this intrinsic trogocytosis is also blocked by Cytochalasin D, a commonly used compound that inhibits the formation of immune synapses. These studies suggest that immune synapse formation is a general mechanism of trogocytosis, and blocking immune synapse formation can interfere with trogocytosis in CAR immune cell therapy, but it can also block trogocytosis in normal immune processes, with unknown benefits or disadvantages.

CAR molecular affinity

CAR molecular affinity is a key indicator related to the safety and efficacy of CAR immunotherapy. High-affinity CAR molecules may produce unnecessary non-targeted or non-tumor binding, that is, binding to normal cells with low antigen expression, which will trigger non-targeted toxicity and other adverse effects, while low-affinity CAR molecules may not bind enough target antigens to activate downstream signals [60]. Now, the influence of CAR molecular affinity on trogocytosis has been determined. High-affinity CAR molecules mediate more target antigen transfer, while low-affinity CD19 CAR-T showed limited trogocytosis in the leukemia mouse model, and no significant reduction of CD19 on the tumor cell surface occurred [61]. The weak affinity has greater sensitivity to antigen density, while other CAR molecular components outside the antigen-binding domain, such as co-stimulatory and intracellular activation domains, do not affect trogocytosis directly [62]. Therefore, a CAR structure with a suitable low affinity could provide a more effective treatment [63].

Regulatory mechanism

The regulatory mechanism of Trogocytosis in CAR-T/NK therapy remains unclear, and recent studies have revealed a mechanism of trogocytosis driven by tumor-derived factors. Cholesterol-25-hydroxylase (CH25H) is a negative regulator of cell membrane cholesterol and plays an important role in key properties such as cell membrane stability and fluidity [64]. Tumor-derived factors (TDFs), such as PGE2 and VEGF, down-regulate the expression of CH25H by activating the Activation of transcription factor 3 (ATF3), promoting trogocytosis between CAR-T cells and tumor cells, while also limiting the viability and anti-tumor activity of CAR-T cells [65].

Functional changes of CAR immune cells after trogocytosis

The direct effect of trogocytosis in CAR-T/NK is that the expression of tumor surface antigen is down-regulated, the passive acquisition of tumor antigen and the expression of CAR in CAR-T/NK cells are decreased, and a series of adverse therapeutic reactions are secondary (Fig. 3A).

Fig. 3.

Trogocytosis mediated impairment of CAR-T/NK dysfunction and its process. A. The direct result of trogocytosis is the loss of antigen on the tumor surface and the decreased expression of CAR molecules, while the production of target-antigen positive CAR-T/NK cells (CAR-T/NK^TROG), which leads to the fratricide of CAR-T/NK cells. Ultimately, the depletion phenotype and metabolic dysregulation of CAR-T/NK cells are triggered, providing a window of opportunity for tumor immune escape. B. Trogocytosis mediated CAR molecular transfer leads to CAR loss on the CAR-T/NK surface and antigen masking of tumor cells, resulting in CAR-T dysfunction and tumor immune escape, which is an important short-term resistance mechanism

CAR-T/NK fratricide and subsequent exhaustion phenotypes

The results of antigen transfer mediated by trogocytosis during normal immune activity are not always positive or negative, but vary according to antigen and immune cell type [66]. For example, NK cells can acquire PD-1 from tumor cell expression via signaling lymphocyte activation molecule (SLAM)-mediated trogocytosis, which results in impaired NK cell anti-tumor activity [67]. Conversely, trogocytosis mediated NK cell transfer of kinase receptor TYRO3 enhanced NK cell activation, cytotoxicity and IFN-γ secretion [68]. The functional impact of trogocytosis mediated antigen transfer in CAR-T therapy has not been elucidated, but the overall effect is that CAR-T cell viability is inversely correlated with the level of trogocytosis mediated antigen transfer [47]. In the study of CAR-NK cells, it was found that the membrane transfer of CD19 promoted the accumulation of IFN-γ and the expression of CD107a in the early stage of CAR-NK cells, but ultimately did not lead to sustained anti-tumor activity [48]. In fact, researchers are more concerned that this antigen transfer provides a target for CAR molecules to mediate the fratricide between CAR-T/NK cells.

When CAR-T/NK cells acquire tumor antigens through trogocytosis (CAR-T/NK^TOGO), they passively become the target of CAR-T/NK. After the fratricide event, the surviving CAR-T/NK^TOGO cells enter a state of exhaustion and decrease in activity, which may be one of the reasons for the recurrence of target antigen-positive tumors in the presence of CAR-T/NK [69]. In fact, intercellular fratricide of CAR-T cells has been reported for a long time, which is based on the co-expression of target antigens on the surface of CAR-T cells and malignant cells, such as CD70, CD26, CD44v6. A recent study revealed the main mechanism of CAR-T cell survival by expressing target genes. It has been found that cannibalistic CAR-T cells are characterized by a exhaustion phenotype (such as TIM3 and LAG3) and a higher clonal expansion status, and they can evade T cell recognition by down-regulating their own HLA expression. While STAT1 and interferon regulatory factors 1 (IRF1) are key regulators regulating the exhaustion, immune escape and immune activation of cannibalistic CAR-T cells, and their down-regulation contributes to the survival and function of cannibalistic CAR-T cells [70].

Metabolic dysregulation and impaired persistence in vivo

In addition to depletion phenotype and impaired viability, trogocytosis is also responsible for metabolic dysregulation in CAR-NK cells, manifested by a substantial decrease in glycolytic capacity [48]. Furthermore, the emergence of trogocytosis in vivo therapy is rapid, independent of initial tumor burden, and accumulates over the course of treatment, resulting in a reduced number of CAR-NK cells, impaired viability, and a depleted phenotype, while reducing the level of target antigens on the tumor cell surface. Analysis of 11 patients with lymphoid malignancies treated with CD19 CAR-NK further supports the adverse impact of trogocytosis on CAR-NK therapy under clinical conditions, as reflected in decreased CD19 expression in B cells from patients associated with acquired CD19 expression on CAR-NK and a higher probability of recurrence [48].

New discovery: CAR molecules on the surface of tumors

As early as 2018, Marco Ruella et al. [71] reported a special case of relapse after CD19 CAR-T therapy. Different from the common causes of relapse such as antigen loss, the CAR molecule was accidentally introduced into cancerous B cells during the production of CAR-T products, and the patient relapsed 261 days after receiving CAR-T therapy, producing a large number of CAR-positive CD19-negative tumor cells. Despite targeted measures, the CAR cancer cells ultimately claimed this patient’s life. Mechanistic studies have found that anti-CD19 CAR molecules form a complex with CD19 on the surface of B-ALL cells, which is called epitope cis-binding, which can prevent the recognition of CD19 CAR-T and confer resistance to CD19 CAR-T. This case not only shows the importance of safe production and quality control of CAR-T cells but also illustrates the dangerous consequences of the acquisition of CAR molecules by tumor cells: the only edited CAR cancer cells eventually exploded in the patient, with the most severe consequences.

Recent studies have identified a novel mechanism of CAR molecular acquisition in tumor cells based on trogocytosis (Fig. 3B). Zhai et al. [49] reported a trogocytosis process influenced by cholesterol metabolism in tumor cells: During the formation of immune synapse between CAR-T and tumor cells, tumor cells are able to acquire CAR molecules and “express” them on the cell surface, which leads to the loss of CAR molecules in CAR-T cells, impaired activity and antigen masking, forming drug resistance against CAR-T cells. Different from the permanent acquisition of CAR molecules in the above cases, the transfer of CAR molecules caused by trogocytosis is reversible and transient, but this short-term resistance provides a window for long-term relapse, supporting the formation of drug resistance at the tumor transcriptome and genomic level, and partially explaining the reason why persistent CAR-T/NK cells cannot prevent long-term relapse.

Intervention methods of trogocytosis in CAR-T/NK therapy

Based on the direct results of a series of preclinical studies and the actual situation encountered in clinical studies, it can be determined that trogocytosis is an important mechanism to hinder CAR-T/NK cell therapy, and blocking trogocytosis can further release the therapeutic potential of CAR-T/NK. Now, several approaches have been proposed to overcome the short-term drug resistance and CAR-T/NK cell dysfunction induced by trogocytosis (Fig. 4).

Fig. 4.

Summary of methods for interfering trogocytosis in CAR immunotherapy

Low-affinity CAR

In the whole research progress of CAR immunotherapy, the structural design of CAR has been the focus [72]. The basic structure of CAR includes the targeted antigen-binding domain, hinge domain, transmembrane domain and intracellular co-stimulatory domain. The addition and modification of intracellular co-stimulatory domain promote the clinical transformation of CAR-T cell therapy, and realize the stable activation of engineered T cells [10]. The hinge and transmembrane domains are less well studied but have also been identified to be associated with the activation of CAR immune cells. The design of antigen-binding domains is usually determined by the therapeutic target, and the scFv is generally designed to correspond to cell surface markers. For example, in leukemia and myeloma, CD19 and CD20 are commonly expressed surface markers on the surface of cancer cells, and anti-CD19 or CD20 scFv is designed as an antigen-binding domain. However, it is now found that in addition to targeting, other properties of scFv, such as affinity, also strongly affect the outcome of treatment [73]. The affinity of scFv accounts for receptor-ligand interactions that fundamentally determine the function of the CAR molecule [74]. In previous studies, CAR molecular affinity was found to be an important safety indicator. Due to the low expression of target antigen on the surface of normal cells, high-affinity CAR-T cells tend to attack these normal cells, while low-affinity CAR-T cells retain the cytotoxic function on tumor cells with high expression of target antigen without toxicity to normal tissues, which encouraged researchers to develop low-affinity CAR molecules [61, 63].

Now, It was found that the affinity of CD19 with CD229 CAR molecule was not necessarily related to the cytotoxicity, but was related to trogocytosis [75]. Compared with high-affinity CAR-T cells, low-affinity CAR-T cells exhibited reduced trogocytosis significantly and maintained higher anti-tumor activity, while high-affinity CAR-T cells underwent more apoptosis and showed lower expansion ability [75]. Results from animal experiments showed that low-affinity CAR-T cells had better durability in vivo, which may be related to reduced trogocytosis, physiological receptor signaling and stem-like phenotype [61]. These results highlight the potential applications of reduced affinity in CAR-T/NK therapy, including reduced trogocytosis and T-cell depletion and increased selectivity, but the expanded antigen profile needs to be carefully reviewed to determine the broad usability of this strategy.

Dual CAR system

The concept of combined CAR-T therapy has been proposed, initially with the aim of combating immune escape caused by antigen loss, so dual-target combined CAR-T therapy was evaluated and tested. The results of an updated clinical trial of CD22 CAR-T and CD19 CAR-T therapy in patients with relapsed or refractory leukemia support that sequential therapy of CAR-T cells with different targets can rescue patients at risk of relapse. In this study, the overall response rate was 98%, and the event-free survival rate at 18 months was 78% [76]. In addition, the co-administration of CD19 CAR-T and CD22 CAR-T also showed promising effects, with 99% of 194 patients with refractory leukemia or relapsed achieving complete remission and a 12-month event-free survival rate of 73.5% [77].

In addition to the dual-target, the fitness between the costimulatory characteristics of different CAR molecules and the density of the target antigen has also been shown to be associated with therapeutic efficacy. Combination of CAR-T therapy with different costimulatory signal domains can mitigate the influence of trogocytosis to some extent. The combination of CD19 CAR-T containing 4-1BB or CD28 produced a stronger killing effect than a simple additive effect. It was found that CD19 CAR(4-1BB) and CD19 CAR(CD28) produced a synergistic effect as the density of CD19 antigen decreased [47]. Furthermore, dual targeting is more effective in preventing tumor immune escape than continuous infusion of a single product [76, 78]. Further studies have found that the combination of CD19 CAR (4-1BB) and CD22 CAR (CD28) is superior to CD19 CAR (4-1BB) and CD22 CAR (4-1BB) or other combinations, although these combination strategies can inhibit tumor recurrence [48]. However, when the expression of CD19 antigen is reduced to a certain extent, CD19 CAR-T therapy can no longer be rescued by co-targeting low-density CD22 targets. This demonstrates the importance of adapting the CAR co-stimulatory features to the density of the target antigen. In conclusion, when trogocytosis occurs, treatment with the dual CAR system objectively blocks this effect, because when one tumor-cell surface antigen is lost by trogocytosis, CAR immune cells targeting the other antigen can still function. At the same time, this short-term resistance overcoming may be one of the key working mechanisms of the dual CAR system.

Design of inhibitory CAR for self recognition

The balance of activating and inhibitory signals determine NK cell-mediated cytotoxicity [79]. In general, in CAR immunotherapy, the purpose of CAR molecule expression is to bind to tumor cells or other abnormal cells expressing specific antigens through the antigen-binding domain of CAR molecules, and signal transduction to the intracellular domain containing immunoreceptor tyrosine-based activation motif (ITAM) to achieve direct activation of immune cells [80]. This activation mode avoids the complicated in vivo immune process and the limitation of immune examination, and achieves the efficient clearance of abnormal cells or tissues. But the development of inhibitory intracellular domains is also valuable. For trogocytosis during CAR immunotherapy, the development of an inhibitory intracellular domain-containing CAR construct (iCAR) targeting specific markers of NK cells may provide an inhibitory signal when CAR-NK cells attack CAR-NK^TOGO to prevent the occurrence of frication, even though these CAR-NK^TOGO cells have tumor antigens on their surface [48]. The concern was whether the inhibitory signal after iCAR activation would interfere with the activity of CAR-NK. It was reassuring that iCAR targeting SLAMF7 successfully interfered with trogocytosis and retained the anti-tumor activity of CAR-NK [48]. The realization of this idea suggests the potential application of inhibitory CAR in CAR immunotherapy, including the prevention of various forms of mediated CAR immune cell fratricide, trogocytosis, the regulation of the balance between activation and inhibitory signals.

CAR enhancement strategy based on key regulatory genes

Trogocytosis is an evolutionarily conserved process, and achieving immune escape through evolutionarily conserved processes is a common trick of tumor cells [81]. But how trogocytosis is regulated in CAR-T/NK therapy remains unclear. Downregulation of CH25H by ATF3, driven by TDFs, has been identified as a key regulatory mechanism for trogocytosis in immunotherapy. Therefore, interference targeting ATF3 and CH25H can affect trogocytosis, including downregulation of ATF3 and up-regulation of CH25H [65]. TAK981, a small ubiquitin-like modifier (SUMO)-ylation inhibitor, was able to restore CH25H expression levels of CTLS exposed to TDFs, effectively blocking trogocytosis, inhibiting tumor growth, and enhancing the efficacy of immunotherapy. The mechanism is unclear, as this SUMO inhibition is not targeted, although ATF3 downregulation and ATF3-dependent CH25H upregulation were observed after treatment. In addition, as an important protein post-translational modification activity, SUMO can regulate various aspects of protein localization, activity and interaction, so the toxicity management of SUMO-ylation inhibitors will be a focus area for future research [82]. Another strategy that has been validated is CH25H coexpression in CAR-T cells. CAR-T cells co-expressing CH25H exhibited enhanced antitumor activity, mitigated trogocytosis, and exhibited an improved exhaustion state [65]. These results suggest that the regulation of ATF3-CH25H axis is an effective measure to block trogocytosis and improve the activity and exhaustion state of CAR-T cells.

Dynamic regulation of CAR molecule expression

A cytotoxic T lymphocyte-associated protein 4 (CTLA-4)-based CAR design has been proposed to inhibit the trogocytosis of CAR-T cells [83]. CTLA-4 is an immune checkpoint molecule with a high degree of endocytic properties and continuously circulates both on and inside the cell surface, which results in limited surface expression [84]. In this process, CTLA-4 cytoplasmic tail (CCT) regulates the surface availability of CTLA-4 to control the optimal activation of T cells [85]. CAR-CCT constructed by the fusion of CCT and the C-terminal of CAR molecule can produce continuous endocytosis, recycling and degradation process, and ultimately reduce the molecular expression of CAR and the secretion of pro-inflammatory factors, which directly reduces trogocytosis. It is worth to say that this process has a quantitative dependence of CCT fusion. However, the anti-tumor effect of CAR-CCT did not decrease and showed long-lasting anti-tumor activity in vivo, possibly due to the compensation of persistence and enhanced central memory differentiation triggered by reduced trogocytosis and CCT modification. This study provides a way to regulate the expression of CAR molecules dynamically through tail fusion of CCT, which has the potential to be more than just trogocytosis and can be used in combination with the inhibitory CAR molecules mentioned above [83]. In addition, antigen expression is often heterogeneous in clinical patients, and expression levels vary from patient to patient. This dynamic expression of CAR molecules controlled by CCT number can provide antigen density adaptive CAR molecule levels.

Other potential methods

In addition to these approaches, there are novel CAR designs that, although not evaluated, may objectively lead to the blockade of the trogocytosis effect. For example, a strategy to prevent or reverse T-cell exhaustion by transiently resting CAR cells through epigenetic regulation was proposed and was successful, with transient loss of CAR molecule expression restoring effector function to T cells that had been plunged into exhaustion [86]. This approach may objectively block the occurrence of trogocytosis and rescue CAR-T/NK^TROG, because trogocytosis is a rapid, transient and reversible process. Similarly, the fusion of CAR molecules with ligand-induced degradation domain can precisely regulate the degradation of CAR molecules through small molecule ligands, and achieve the loss of CAR molecules in a spatiotemporal sense, which may have a positive significance for trogocytosis [87].

Conclusion

Bidirectional trogocytosis between CAR immune cells and tumor cells is an important general mechanism of short-term drug resistance, and provides an opportunity for long-term relapse at the transcriptome or proteomic levels, which is an important reason for tumor recurrence in the presence of CAR-T/NK. In addition, considering the widespread existence of trogocytosis effect in human cells, it is also possible that CAR immune cell therapy for other diseases may be interfered by trogocytosis, such as CAR-T therapy for autoimmune diseases, which has achieved substantial progress. Therefore, trogocytosis should be a key parameter to be valued and evaluated in the design of novel CAR immune cells. Now, benefit from the progress in mechanism research, some methods for trogocytosis have been proposed, including designing low-affinity CAR structures, dual CAR cell therapy, and designing CAR based on inhibitory signals, etc. It is expected that these approaches will become an important reference for CAR immune cell therapy, and further research on trogocytosis mechanisms will contribute to better understand the cell biology and product development of CAR immune cells.

Abbreviations

ATF3

Activation of transcription factor 3

BCMA

B cell maturation antigen

CAR

Chimeric antigen receptor

CAR-M

Chimeric antigen receptor macrophage

CAR-T/NK^TROG

CAR-T/NK cells with tumor antigens were obtained by trogocytosis

CCT

CTLA-4 cytoplasmic tail

CH25H

Cholesterol-25-hydroxylase

CTLA-4

Cytotoxic T lymphocyte-associated protein 4

iCAR

Inhibitory chimeric antigen receptor

IRF1

Interferon regulatory factors 1

ITAM

Immunoreceptor tyrosine-based activation motif

MSLN

Mesothelin

SLAM

Signaling lymphocyte activation molecule

SUMO

Small ubiquitin-like modifier

TDF

Tumor-derived factor

Author contributions

C (Writing - original draft, Visualization); X (Writing - original draft); Z (Supervision, Writing - review & editing); Q (Supervision, Writing e review & editing); Q (Supervision, Writing e review & editing); L (Conceptualization, Supervision, Writing e review & editing); T (Conceptualization, Supervision, Writing e review & editing); All authors reviewed the manuscript.

Funding

This work was supported by the Basic and Clinical Cooperative Research Program of Anhui Medical University Incubation Project for The Third Affiliated Hospital (2022sfy017).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

All authors give consent to publish this manuscript.

Competing interests

The authors declare no competing interests.