17.3 Synthetic Biology Approaches to Cellular Immunotherapy Engineering 355sequential antigen encounter. Such computational capability offers a means to
increase the safety and efficacy of adoptive T-cell therapy by addressing critical
clinical challenges, including imperfect targeting specificity and vulnerability
to antigen escape (i.e., a process by which diseased cells escape T-cell detection
by downregulating the expression of targeted antigens).
For example, three recent clinical trials reported that 38–100% of patient
relapses after CD19 CAR-T cell therapy were characterized by the loss of CD19
expression [28, 63, 64]. To address the problem of antigen escape, bispecific
OR-gate CARs that incorporate two scFv domains have been developed. T cells
armed with OR-gate CARs can respond to either of two distinct antigen inputs,
thus reducing the probability that a tumor cell can successfully escape detection
via mutational loss of antigen expression [65, 66] (Figure 17.2b). This principle
has been applied to generate an optimized CD19/CD20 bispecific CAR that ena-
bles cytotoxic T cells to effectively eliminate cancerous B cells that have lost
CD19 expression [66, 67]. Specifically, T cells expressing the bispecific CAR are
able to not only eradicate established lymphoma in mice but also prevent tumor
relapse, whereas animals treated with conventional, single-input CD19 CAR T
cells succumb to cancer recurrence caused by antigen escape [66, 67]. Additional
combinations such as CD19/CD22 are also under active preclinical evaluation
[68], and they promise to significantly increase the efficacy of CAR-T cell therapy
against heterogeneous and/or genetically unstable tumors.
Figure 17.2 CARs redirect T-cell specificity toward tumor targets. (a) Schematic of first-, second-,
and third-generation CARs. The single-chain variable fragment (scFv) derived from a tumor-
antigen-specific antibody serves as the extracellular sensing domain, and the cytoplasmic tail of
the CD3ζ chain serves as the intracellular signaling domain of the CAR. In second- and third-
generation CARs, one or two costimulatory domains such as CD28 and 4-1BB are directly fused
to the CD3ζ chain to enhance T-cell signaling. (b) Schematic of single-chain, bispecific OR-gate
CARs. T cells expressing an OR-gate signal processing system can kill any target cell that
expresses either antigen A or antigen B. (c) Schematic of an AND-NOT-gate CAR pair. The first
receptor is a conventional CAR that targets antigen A. The second is a chimeric inhibitory
receptor (iCAR) that targets antigen B and contains the cytoplasmic domain of an inhibitory
receptor (e.g., PD-1 or CTLA-4). Presence of antigen A triggers CAR signaling, while presence of
antigen B triggers iCAR signaling. The inhibitory function of the iCAR overrides any activation
signal that may result from the conventional CAR, thus executing A-NOT-B signal computation.
(d) Schematic of an AND-gate CAR pair. The first receptor is a conventional first-generation CAR
that targets antigen A and contains only the CD3ζ chain without costimulatory signals. The
second is a chimeric costimulatory receptor that targets antigen B and contains both CD28 and
4-1BB costimulatory signals but no CD3ζ chain. Both antigens must be present to trigger a
sufficiently robust T-cell response to execute therapeutic function. (e) Schematic of a “remote-
controlled” CAR system. Here, the CAR protein is split into two parts, with the first fragment
being a conventional CAR that contains the FK506 binding protein (FKBP) instead of the CD3ζ
chain at the C-terminus. The second fragment consists of a membrane-tethered CD3ζ chain
fused to the FKBP-rapamycin binding (FRB). Presence of a rapamycin analog (rapalog) molecule
triggers dimerization between FKBP and FRB, thereby reconstituting a full CAR protein and
enabling CAR signaling in response to antigen binding. (f ) Schematic of a synthetic Notch
(synNotch) receptor-regulated CAR expression system. Upon binding to antigen A, the synNotch
receptor releases a TF, which translocates to the nucleus and triggers CAR expression from a
cognate promoter. This CAR molecule is subsequently able to trigger T-cell activation upon
binding to antigen B, resulting in AND-gate signal computation in a sequential manner.