Background

Chimeric antigen receptor (CAR) T-cell therapy is an engineered cellular cancer therapy that reprograms a patient’s T cells to target a specific protein, mimicking natural T-cell receptor (TCR) function but redirected toward antigens found on the surface of cancer cells. CAR-T cells have emerged as a highly successful tool for providing robust therapeutic responses against refractory or relapsed hematological cancers, with ongoing research across many domains of immunology to extend the range and efficacy of CAR-T treatments for solid tumor malignancies [1–3].

Many advancements have been made in understanding the novel biology of CAR-modified T cells; however, CAR-T-cell screening models still struggle to choose candidate binding elements based on in vitro assessments of CAR-T activation and cytotoxicity assays. Whilst activation markers, such as CD69, are well-documented using CAR-Jurkat and primary CAR-T-cell screening methods, it is unclear how well these markers can predict in vivo efficacy of the CAR-T cells [4,5]. Recently, cellular avidity was recognized as a property of CAR-expressing cells that varies between different CAR constructs and correlates with important functional properties such as CAR expansion, CAR-activated trogocytosis, and CAR-T exhaustion [6–9].

Cellular avidity is defined by the total sum of cell-to-cell interactions, encompassing the whole immunological synapse and many cellular adhesion processes [10]. Given that cell binding is a near-immediate consequence of CAR signaling and proximal T-cell response processes, understanding the nature of this interaction and how it differs between standard TCR interaction vs. CARs is an important area of research. Investigating how cellular avidity changes with different CAR binding domains or structural designs is also vital for improving CAR discovery and development. It is important to note that while the specific properties of an antibody used to create a CAR impact the overall cellular avidity, the affinity and avidity of an antibody are entirely distinct from the complete cell-to-cell interaction that defines “cellular avidity”.

For soluble proteins such as antibodies, avidity is defined as the overall accumulated strength of interaction through multiple interactions, but it can vary widely depending on the nature of the target antigen and the specific conditions of the assay. The measured avidity of an antibody is usually reported as the half-maximal binding concentration observed over a titrated binding assay, this property being referred to as the apparent affinity (K_D). In contrast to antibodies, using the assay described herein, we have observed that the half-maximal binding of CAR-expressing cells over a titrated range of target cell concentration does not vary between CAR proteins; rather, we find that CAR cells consistently reach a point of saturated maximal binding, wherein a varying proportion of CAR-expressing cells will form a strong bond with target cells (we refer to this as a CAR-target doublet). The strength of this interaction, while not encompassing the entirety of the kinetics, approximates what is commonly called cellular avidity. This observation of a varying binding rate between CAR-expressing and target cells has similarly been reported using an assay based on a resonance force ramp microscopy using the Lumicks zMovi device [6], which directly tests the strength of interaction by attempting to measure the force at which CAR-target pairs are dissociated. We find that the rate of CAR-target doublet formation increases with higher target cell density, eventually reaching a saturation point. Furthermore, we find the rate of doublet formation varies between different CARs [11] or CAR designs [5]. While this assay does not directly test the strength of interaction and therefore is only an approximation of overall cellular avidity, we believe the unique target binding properties for each CAR serve as valuable metrics for CAR screening and assessment. As such, this assay could be used as a replacement to, or in conjunction with, direct avidity assays for optimal CAR selection.

Here, we provide a complete description of our cellular binding assay that serves as a quick, high-throughput addition to CAR-T screening workflows using a standard flow cytometer. Through the quantification of cell/target doublets formed within CAR Jurkat target cell co-cultures over a wide range of effector-to-target ratios, the unique cell binding properties associated with each CAR can be determined. This assay provides consistent results and can be employed with both solid tumors (see validation of protocol) and hematological tumor models.