BetaEntity Annotation Prototype
← Back to diseases

Annotated abstract

205 Single-cell functional profiling reveals serial killing dynamics of CD33-targeting adapter UniCAR-T cells

jitc · 2025-11-04 · canonical JSON source

3 visible annotations · policy: published · automated confidence ≥ 75.00%

Document resource

Background CAR-T cells display marked phenotypic and functional heterogeneity, which can critically influence therapeutic efficacy. However, dissecting the single-cell dynamics underlying this diversity remains technically challenging. Adapter CAR-T platforms, such as the previously described UniCAR-T system, offer great promise for enhancing CAR-T safety and efficacy as the function of UniCAR-T cells can be regulated via the presence of a targeting module (TM). Nevertheless, little is known about the single-cell functional dynamics and serial killing capabilities of UniCAR-T cells.Methods We previously developed a novel microfluidic technology for the generation of miniaturized co-cultures in a microfluidic array containing over 19,000 microwells. In this study, we leverage this high-throughput platform to image, track and analyze thousands of UniCAR-T-tumor cell co-cultures at single-cell resolution. Through a combination of robotics, time-lapse imaging and AI-driven image processing, we analyze the tumor-targeting dynamics of individual UniCAR-T cells redirected with increasing concentrations of an anti-CD33 TM against acute myeloid leukemia (HL-60) cells.Results Tracking the killing capabilities of thousands of UniCAR-T cells redirected via an anti-CD33 TM revealed a substantial heterogeneity within and across donor sources. Through single-cell dynamic profiling, UniCAR-T cells were classified into distinct functional subsets exhibiting varying tumor-killing dynamics. Single-cell analyses identified a subpopulation of highly potent serial killers with enhanced cytotoxic potential. While many UniCAR-T cells were not able to kill any tumor cell, some individual UniCAR-T cells were able to kill over ten tumor cells, unveiling a previously unreported degree of single-cell functional heterogeneity not captured by conventional bulk assays. Analysis of TM-dependent redirection on UniCAR-T cells highlighted its role in sustaining increased serial killing activity, which reached around 20% at the latest time points analyzed. Additionally, we introduced a novel single-cell in vitro potency index, termed Unified Killing Score, which further confirmed the specific activation of redirected UniCAR-T cells at single-cell resolution. Further multiparametric analyses directly linked functional activity to the expression of specific phenotypic markers, such as HLA-DR, also revealing the cytolytic capabilities of specific CD4+ and CD8+ UniCAR-T subsets.Conclusions Altogether, our research provides a detailed single-cell characterization of the tumor-targeting dynamics of adapter UniCAR-T cells. By integrating functional, phenotypic and kinetic readouts across thousands of CAR-T-tumor cell co-cultures, we unveil the serial killing capabilities of UniCAR-T cells, identifying a subset of cells with superior tumor-killing potential. This microfluidics-based approach may offer a valuable platform for optimizing the rational design and development of next-generation CAR-T therapies.