A dual chain chimeric antigen receptor (CAR) in the native antibody format for targeting immune cells towards cancer cells without the need of an scFv.

Adoptive cell therapy with chimeric antigen receptor (CAR)-modified T cells showed remarkable therapeutic efficacy in the treatment of leukaemia/lymphoma. However, the application to a variety of cancer entities is often constricted by the non-availability of a single chain antibody (scFv), which is usually the targeting domain in a CAR, while antibodies in the natural format are often available. To overcome the limitation, we designed a CAR that uses an antibody in its natural configuration for binding. Such CAR consists of two chains, the immunoglobulin light and heavy chain with their constant regions, whereby the heavy chain is anchored to the membrane and linked to an intracellular signalling domain for T-cell activation. The two chains form a stable heterodimer, a so-called dual chain CAR (dcCAR), and bind with high affinity and in a specific manner to their cognate antigen. By specific binding, the dcCAR activates engineered T cells for the release of pro-inflammatory cytokines and for target cell lysis. We provide evidence by three examples that the dcCAR format is universally applicable and thereby broadens the CAR cell therapy towards a larger variety of targets for which an scFv antibody is not available.

T cells redirected by a CD3ζ chimeric antigen receptor can establish self-antigen-specific tumour protection in the long term.

A majority of cancer deaths are because of an uncontrolled relapse of the disease despite initial remission after therapy, asking for strategies to control tumour cells in the long term. Adoptive therapy with chimeric antigen receptor (CAR)-redirected T cells showed promising success in primary tumour elimination; the capacity of such engineered T cells to establish enduring tumour protection is currently a matter of discussion, in particular as most targeted 'tumour-associated antigens' are self-antigens. To address the issue in a clinically relevant model that closely mimics the human situation, we recorded rejection of carcinoembryonic antigen (CEA)-positive pancreatic tumours in the CEA transgenic mouse that expressed CEA as self-antigen in healthy cells of the gastrointestinal tract. Adoptive therapy with CD8(+) T cells, which were redirected by a CEA-specific, low-affinity CAR with CD3ζ endodomain, eliminated CEA(+) tumours in a primary response; cured mice produced an efficient recall response in the long term towards CEA(+) tumour cells upon rechallenge. Secondary tumour rejection was CEA specific, mediated by engineered T cells and did not require host T cells. No toxicity towards healthy tissues with CEA expression was recorded. Data indicate that adoptive therapy with engineered T cells can establish self-antigen-specific tumour protection in the long term without autoimmunity.

The weal and woe of costimulation in the adoptive therapy of cancer with chimeric antigen receptor (CAR)-redirected T cells.

Adoptive cell therapy has shown impressive efficacy to combat cancer in early phase clinical trials, in particular when T cells engineered to specifically target tumor cells were applied. The patient's T cells are genetically equipped with a chimeric antigen receptor (CAR) which allows them to be redirected in a predefined manner towards virtually any target; by using an antibody-derived domain for binding, CAR T cells can be redirected in a major histocompatibility complex (MHC) dependent and independent fashion. The CAR also provides the stimuli required to induce and maintain T cell activation. Recent clinical data sustain the notion that strong costimulation in conjunction with the primary activation signal is crucial for lasting therapeutic efficacy of CAR T cells. However, costimulation is a double-edged sword and the impact of the individual costimuli to optimize T cell activation is still under debate; some general rules are emerging. The review summarizes how costimulation modulates, improves and prolongs the redirected anti-tumor T cell response and how the same costimulatory signals may contribute to unintended side effects including "cytokine storm" and T cell repression. Upcoming strategies to break the activation/repression circle by using CAR's with modified costimulatory signals are also discussed.

CD28 cosignalling does not affect the activation threshold in a chimeric antigen receptor-redirected T-cell attack.

Adoptive immunotherapy of cancer using chimeric antigen receptor (CAR)-engineered T cells with redirected specificity showed efficacy in recent trials. In preclinical models, 'second-generation' CARs with CD28 costimulatory domain in addition to CD3ζ performed superior in redirecting T-cell effector functions and survival. Whereas CD28 costimulation sustains physiological T-cell receptor (TCR)-CD3 activation of naïve T cells, the impact of CD28 cosignalling on the threshold of CAR-mediated activation of pre-stimulated T cells without B7-CD28 recruitment remained unclear. Using CARs of different binding affinities, but same epitope specificity, we demonstrate that CD28 cosignalling neither lowered the antigen threshold nor the binding affinity for redirected T-cell activation. 'Affinity ceiling' above which increase in affinity does not increase T-cell activation was not altered. Accordingly, redirected tumor cell killing depended on the binding affinity but was likewise effective for CD3ζ and CD28-CD3ζ CARs. In contrast to CD3ζ, CD28-CD3ζ CAR-driven activation was not increased further by CD28-B7 engagement. However, CD28 cosignalling, which is required for interleukin-2 induction could not be replaced by high-affinity CD3ζ CAR binding or high-density antigen engagement. We conclude that CD28 CAR cosignalling does not alter the activation threshold but redirects T-cell effector functions.

Adoptive immunotherapy with genetically engineered T cells: modification of the IgG1 Fc 'spacer' domain in the extracellular moiety of chimeric antigen receptors avoids 'off-target' activation and unintended initiation of an innate immune response.

Chimeric antigen receptors (CARs, immunoreceptors) are frequently used to redirect T cells with pre-defined specificity, in particular towards tumour cells for use in adoptive immunotherapy of malignant diseases. Specific targeting is mediated by an extracellularly located antibody-derived binding domain, which is joined to the transmembrane and intracellular CD3ζ moiety for T-cell activation. Stable CAR expression in T cells, however, requires a spacer domain interposed between the binding and the transmembrane domain and which is commonly the constant IgG1 Fc domain. We here revealed that CARs with Fc spacer domain bind to IgG Fc gamma receptors (FcγRs), thereby unintentionally activating innate immune cells, including monocytes and natural killer (NK) cells, which consequently secrete high amounts of pro-inflammatory cytokines. Engineered T cells, on the other hand, are likewise activated by FcγR binding resulting in cytokine secretion and lysis of monocytes and NK cells independently of the redirected specificity. To reduce FcγR binding, we modified the spacer domain without affecting CAR expression and antigen binding. Engineered with the modified CAR, T cells are not activated in presence of FcγR(+) cells, thereby minimizing the risk of off-target activation while preserving their redirected targeting specificity.

Redirecting human CD4+CD25+ regulatory T cells from the peripheral blood with pre-defined target specificity.

Recent insight into the balance of self-tolerance and auto-aggression has raised interest in using human regulatory T (Treg) cells for adoptive immunotherapy of unlimited autoimmune diseases including type-1 diabetes, rhematoid arthritis and multiple sclerosis. The therapeutic use of Treg cells, however, is so far hampered by the inefficiency of current protocols in making them accessible for genetic manipulations. We report here that TCR/CD3 stimulation that is accompanied by extensive CD28 costimulation makes human Treg cells susceptible to retroviral gene transfer ex vivo while preserving their properties in vitro and in vivo. To show the power of genetic manipulation of human Treg cells, we engineered 'designer Treg cells' by retroviral expression of a chimeric immunoreceptor with defined specificity, which activates Treg cells in a ligand-dependent manner to proliferate, to secrete high amounts of interleukin-10 and to repress an ongoing cytolytic T-cell response in vivo. The procedure in genetically modifying human Treg cells ex vivo will open a panel of applications for their use in the adoptive therapy of deregulated immune responses.