Universal CAR 2.0 to overcome current limitations in CAR therapy

Abstract

Chimeric antigen receptor (CAR) T cell therapy has effectively complemented the treatment of advanced relapsed and refractory hematological cancers. The remarkable achievements of CD19- and BCMA-CAR T therapies have raised high expectations within the fields of hematology and oncology. These groundbreaking successes are propelling a collective aspiration to extend the reach of CAR therapies beyond B-lineage malignancies. Advanced CAR technologies have created a momentum to surmount the limitations of conventional CAR concepts. Most importantly, innovations that enable combinatorial targeting to address target antigen heterogeneity, using versatile adapter CAR concepts in conjunction with recent transformative next-generation CAR design, offer the promise to overcome both the bottleneck associated with CAR manufacturing and patient-individualized treatment regimens. In this comprehensive review, we delineate the fundamental prerequisites, navigate through pivotal challenges, and elucidate strategic approaches, all aimed at paving the way for the future establishment of multitargeted immunotherapies using universal CAR technologies.

Keywords: CAR (chimeric antigen receptor) T-cell therapy; antibody therapies; cancer immune cell therapy; iPSC (induced pluripotent stem cell); individualized cancer therapy.

Figure 1

Physicochemical signal conversion of CAR receptors. CARs are (A) mechanoreceptors that allow for the transmission of physical tension force to convert into complex chemical signals from the outside of the cell into changing the activation state and downstream functions of the cell including immediate responses, such as the formation of a cytolytic synapse and migration or intermediate and long-term adaptions by changes in the gene expression. (B) The mechanical force on CAR expressing effector cells is mediated by the binding interaction of the CAR recognition domain and the antigen leading to a conformational change of the cytoplasmic signaling domains. These steric adaptions result in the accessibility of phosphorylation sites and the cognate alignment of signaling proteins that initiate a downstream signaling cascade which is based on enzymatic phosphorylation steps of subsequent signaling proteins in a defined order (13). Calcium serves as an important second messenger in the signaling process. Piezo1 calcium channels are opened by tension forces to the cell membrane by active deformation of the cytoskeleton via actin filaments, allowing calcium influx into the cytoplasm. Especially, during the formation of the cytolytic synapse Piezo1 dependent calcium influx is required (14). Further, cellular calcium metabolism is tightly regulated by the endoplasmic reticulum (ER). The ER serves as a large calcium storage filled with calcium by the calcium release activated channel (CRAC) (15). Upon calcium release from the ER and binding thereof to calcineurin, a phosphatase, is activated. Calcineurin serves as a key modulator of the transcription factor “nuclear factor of activated T cells” (NFAT) and thus serves as a key modulator of T cells in general (16). Dephosphorylation of the nucleus localization signal allows the translocation of NFAT into the nucleus and induce selective gene expression. Dysfunctional CRAC prevents the development of T cells and is the cause of the severe combined immunodeficiency (SCID), a life-threatening inborn immune deficiency (17). (C) The efficient transmission of the physicochemical signal leads to various effector functions in CAR expressing effector cells. In a resting CAR, 1^st cell activation by the CAR receptor is detectable by the expression of activation markers. 2^nd Cytolytic activity in high affinity CARs (KD < 1 nM) is induced at an antigen density of as low as >200 molecules per cell, whereas significant changes to the 3^rd gene expression, cytokine and chemokine secretion require >2,000 molecules per cell (18). Thus, 4^th a strong triple signal of ⁱ⁾ CAR-mediated antigen recognition (CD3ζ), ⁱⁱ⁾ co-stimulation (CD28), and ⁱⁱⁱ⁾ cytokine support (IL2) are required to induce proliferation.

Figure 2

Universal CAR strategies have two meanings. Universal “allogeneic CARs” generated from iPSCs require a series of genetic alterations to enable their successful and safe clinical use. Current concepts (A) include modifications that reduce the likelihood of allorejection by gene disruption of ß2-microglobulin to abrogate the expression of HLA class I structures as well as by gene disruption of the class II MHC transactivator (CIITA) to abrogate HLA II expression (41, 42). To exclude iPSC-derived cells from harm through alemtuzumab, the CAMPATH-1 antigen CD52 is required to be knocked out as it has been done for third-party allogeneic CAR products (43). Alemtuzumab can then be used in the CAR preparative regimen and facilitate longer engraftments of the allogeneic cells. However, alemtuzumab induces long-lasting severe cellular and humoral immune deficiency which attracts serious infectious complications (44). In order to reduce the risk for graft-versus-host disease, the T cell receptor (TcR) expression has to be disrupted and is usually achieved by the genetic knockout of the constant alpha (TRAC) chain of the TcR. Since the TcR is the responsible receptor for alloreactivity in GvHD direction, the TcR-KO is the most effective strategy to silence the primary T cell function (42). Novel strategies to increase the resistance to potential allorejection was introduced by Mo et al in 2021 via an alloimmune defense receptor (ADR) (45). It’s a ligand based (4-1BBL) signal converting receptor by providing a CD3ζ signaling if the allogeneic iPSC derived CAR expressing cell gets in contact with activated immune cells, such as T cells and NK cells that express the co-stimulatory receptor 4-1BB (CD137). The CD3ζ signal increases the resistance to allorejection mechanisms induced by T cells and NK cells. (B) The basic principle of adapter CARs is the indirect targeting of the CAR expressing cells via advanced antibody-dependent cellular cytotoxicity (ADCC). Since the adapter molecule, e.g., an antibody is interchangeable the specificity of the targeting is theoretically unlimited. The structure of the adapter molecule is comprised of three domains with distinct functions. The antigen binding domain corresponds to the primary antibody binding capability to a structure expressed on target cells. The structural domain provides stability to the molecule and supports manufacturability, the connecting module interacts with the CAR receptor and facilitates the highly specific recognition of the adapter molecule only by the CAR expressing cell and no other immune cells. The counterpart is the CAR expressing cell and especially the CAR receptor itself comprised of the cognate connecting module to allow the highly specific recognition and interaction with the adapter molecule (20). Further, the anchoring domain stabilizes the receptor in the cell membrane and the signaling domain provides the cell with the downstream signaling to ignite cellular functions according to the design of the receptor.

Figure 3

Universal CAR configurator. The concept of the Universal CAR Configurator is to design the most effective individual universal CAR for each specific cancer indication. It can be used with autologous immune effector cells or more advanced with iPSC-derived cell products to create “Universal CAR 2.0”. (A) The process begins with a comprehensive analysis of the cancer using state-of-the-art diagnostics. This includes basic microscopy, immunohistochemistry, and the detection of chromosomal aberrations (both structural and numerical) as well as utilizing fluorescence in situ hybridization (FISH). Additionally, targetable mutations and epigenetic changes are identified. Standard diagnostics confirm the diagnosis and guide the initial treatment regimen. Advanced diagnostics are employed to identify patients who could benefit from additional treatment options, either through established therapies or personalized treatments. Established therapies may involve the use of small molecules to inhibit upregulated signaling pathways such as kinases, mTOR, ABL, c-KIT, and FLT3. Moreover, patients with DNA instability due to mismatch repair deficiency, such as those with colorectal cancer, can significantly benefit from immune checkpoint inhibition (226). Further, theranostic approaches with antibody-guided radionuclides, such as Iodine-131 coupled to antibodies are used therapeutically (227). Whole genome sequencing is used to identify targetable mutations in the tumor with patient-individual cancer vaccines (228). Functionally tailored CAR-based therapies require a patient-individual two-step immune profiling. In the first step, the cancer is screened for the expression of relevant genes. Based on the results, in a second step a specific antibody panel is used for spatial proteomics to elucidate the patient’s tumor microenvironment (TME), exemplified as 1^st PDL1 expression, 2^nd TGFß secretion and 3^rd CCL19 chemokine release by the tumor. (B) Therefore, the design of the TME-adjusted patient-individual CAR product requires the incorporation of 1^st a PD1-CD28 signal converting receptor, 2^nd a dominant negative TGFß receptor and 3^rd the introduction of the chemokine receptor CCR7 to support the enrichment of CAR effector cells in and around the tumor. (C) The genetic payload of the relevant additional effector functions may be incorporated into the expression cassette and transferred via stable integration of the transgene using viral, transposon-based or CRISPR-based gene delivery. Additional genetic modifications can be utilized to inactivate immunosuppressive receptors such as PD-1 or CTLA-4 and others. (D) Spatial proteomics is used for the selection of patient-individual antibody panels that are specifically formulated for each patient. Combinatorial targeting is the key obstacle of targeted immunotherapies to overcome the antigen heterogeneity and immune escape evasion mechanisms. (E) Although patients receive individualized therapies, they built on generic key components - the CAR receptor and the adapter molecules. DX, diagnostics. TX, therapeutics. AM, adapter molecule. TGFß, TGF beta. IL15R, IL15 receptor. PD1-CD28, signal converting receptor. DN-TGFßR, dominant negative TGF beta receptor. CCR/CXCR, chemokine receptors. TNFRS, tumor necrosis factor receptor superfamily. CIL7, constitutive IL7 receptor.

Figure 4

Key determinants of CAR signaling. (A) The CAR receptor-target antigen interaction in conventional CAR technologies is primarily determined by the antigen expression. Most target antigens are expressed at lower levels than the CAR receptor on the effector cell population. Thus, in antigen high expressing target cells, the avidity is higher than in antigen low expressing targets and as the CAR signaling is proportional to the CAR receptor-target antigen interaction, it is enhanced in antigen high expressing targets. (B) Besides the target antigen density, in adapter CAR technologies the adapter molecule concentration significantly impacts on the CAR engagement with the target cells. Without any adapter molecule available, there is no CAR engagement possible. With increasing concentrations, the CAR engagement becomes more efficient and reaches an optimum before inhibitory effects start to reduce the CAR-target interaction at supra-optimal adapter molecule concentrations. (C) Synchronic multitargeting, utilizing a combination of adapter molecules at low concentrations, can increase the CAR-target interaction while the blocking effects are reduced. (D) Key determinants of CAR-target interaction are ⁱ⁾ the antigen expression density on the target cells, ⁱⁱ⁾ the CAR receptor expression density on the CAR⁺ effector cell population, and the ⁱⁱⁱ⁾ interplay of the avidity and the affinity (K_D) of the adapter molecule to the targeted antigen as well as the CAR recognition domain, e.g., CAR-scFv to the CAR-target moiety on the adapter molecules. (E) At supra-optimal concentrations, competitive blocking effects reduce the probability of CAR-target interaction and therefore pharmacokinetic as well as pharmacodynamic aspects must be considered optimizing the adapter molecule dosing.