By mapping CAR-T cells one by one, researchers are uncovering the cellular traits that may separate durable cancer responses from relapse.
Review: Mapping clinical CAR-T cells: insights from scRNA-seq. Image Credit: Nemes Laszlo / Shutterstock
In a recent review published in the journal Trends in Molecular Medicine, researchers in the United States synthesized findings from single-cell RNA sequencing (scRNA-seq; n = 44 studies) to identify transcriptional patterns associated with how well a patient responds to CAR-T cell therapy.
Single-Cell Analysis of CAR-T Response
Chimeric antigen receptor T cell (CAR-T) therapy is considered one of the landmark achievements of modern medicine. The therapy involves harvesting a patient’s own immune cells, genetically engineering them to recognize specific tumor markers, and infusing them back into the body to hunt and destroy cancer.
However, long-term evaluations increasingly show that while this approach has led to durable remissions in previously untreatable leukemias and lymphomas, a significant portion of patients still face relapse or resistance. One possible reason is that, until recently, researchers could study these cells only in "bulk," looking at the average behavior of millions of cells at once.
This lack of fine-scale resolution meant that highly effective cells, along with the "exhausted" ones that fail early on, could be obscured from the dataset, potentially leading to incomplete conclusions.
The emergence of single-cell RNA sequencing (scRNA-seq), a method that tags and sequences messenger RNA (mRNA) from individual cells, has provided a high-resolution analytical tool for observing these dynamics at the single-cell level. Understanding these individual cellular journeys may be the key to determining why some therapies persist long-term while others are short-lived.
scRNA-seq Review Scope and Cohort
The present review aimed to address this knowledge gap by synthesizing findings from 44 different studies that utilized scRNA-seq to track CAR-T cells in human patients. The cohort represented a total of 500 patients, primarily those with hematological (blood) malignancies, including B-cell lymphoma (197 patients), acute lymphoblastic leukemia (ALL) (159 patients), multiple myeloma (54 patients), and solid or brain tumors (80 patients).
Study methodologies predominantly comprised 10X Genomics data from infusion products (the "drug" itself), peripheral blood, bone marrow, and other sources. The analyses’ key phenotypes and readouts included: 1. Exhaustion, the loss of cell function over time, 2. Cytotoxicity, the cell's tumor-killing activity, often involves proteins like granzymes and perforin, 3. Memory and stemness, the ability of cells to persist and multiply long-term, and 4. Clonal diversity refers to how many unique types of T cell "families" (clones) survived the treatment process.
Exhaustion, Memory, and Clonal Dynamics
The present review findings suggest that cellular exhaustion is one of the most consistent correlates of poor CAR-T clinical outcomes. Across multiple studies, patients who did not respond to therapy or relapsed early showed higher expression of exhaustion-related genes like LAG3, PDCD1 (PD-1), and HAVCR2 (TIM-3).
The data further identifies a "balancing act" between cytotoxicity and memory, demonstrating that several studies linked better responses with infusion products enriched for memory-like cells, while cytotoxic programs appeared more time- and context-dependent. Specifically, data from five separate studies showed that better responders had a higher proportion of memory cells at the time of infusion.
However, once inside the patient, these memory cells must differentiate into more cytotoxic cells to eliminate the tumor. Overall, results from the review revealed substantial target prevalence, with the majority of sequenced patients (329) receiving CAR-T cells targeting CD19, the most common marker for B-cell cancers.
Furthermore, clonal dynamics investigations showed that some long-term responders had highly expanded persistent CAR-T cell clones, although clonal patterns varied by disease and study. These findings suggest that persistence, clonal diversity, and cell state may all influence long-term remission.
Brain Tumor CAR-T Tracking and Future Directions
Finally, emergent research highlights that while brain tumors are harder to treat due to the blood-brain barrier, early studies of brain tumors, including analyses of cerebrospinal fluid (CSF) and tumor samples, suggest that scRNA-seq can help track CAR-T cell activity and immune remodeling in central nervous system settings. However, the evidence remains preliminary, with limited sample numbers and challenges in obtaining representative tumor samples.
This review is among the first to provide a broad synthesis, showing that scRNA-seq is helping make CAR-T responses more interpretable at single-cell resolution. Its findings highlight crucial hurdles that must be overcome before CAR-T therapy can further advance modern oncological treatment: the high cost of sequencing, the need for specialized bioinformaticians, and the current lack of standardized data across labs.
The authors conclude that as sample sizes grow and technology becomes more affordable, these high-resolution genetic insights could help researchers refine T cells for maximum fitness, potentially bringing the success seen in blood cancers to the much more challenging realm of solid tumors.
Download your PDF copy by clicking here.
New cord blood approach boosts survival in blood disease patients