Apr 6, 2023 / Oncology

A movement rises to guide CAR T-cell therapy development with single cell multiomics

Olivia Habern

Immunotherapies such as chimeric antigen receptor (CAR) T-cell therapies have transformed cancer treatment, offering effective treatment options to some cancer patients, especially those with hematologic cancer. But these therapies still show mixed results due to exhaustion or toxicity and limited efficacy in solid tumors.

Clinical oncologists want to broaden CAR T therapeutic efficacy to new patient populations beyond blood cancers. They are working tirelessly to deeply characterize CAR T products and the tumor-immune landscape pre- and post-treatment, in search of clues to the factors that contribute to therapeutic efficacy or resistance and disease relapse. Life-saving therapeutic improvements may be hidden in a patient’s own cells and tumors, leading researchers to search precious patient samples—blood, bone marrow, excisional biopsies—for answers.

But these unrealized therapies are cloaked by biological complexity. High-resolution, multiomic tools are necessary to understand the complexity of these samples and clarify the layers of cellular biology, including the epigenome, transcriptome, and immune repertoire, that dictate CAR T-cell states and functionality.

And don’t just take our word for it—a growing movement of scientists recognize the need for rigorous, comprehensive approaches to advance CAR T therapy development, as evidenced by recent publications pointing to the value of single cell multiomic profiling technologies for this effort (1–4). Recognizing the need for collaboration and shared knowledge in order to tackle the biggest challenges presented by T cell–based cancer immunotherapy, in 2021, 10x Genomics and the Parker Institute of Cancer Immunotherapy (PICI) came together for a workshop to evaluate the state of T-cell research and identify areas for focused technology development (5).

Clinical oncologists are not alone in searching for the ideal T-cell therapy to advance the goal of long-term, progression-free survival in their patients. And new discoveries are adding momentum to the movement: keep reading to learn how scientists have leveraged single cell multiomics to uncover new predictive biomarkers for neurotoxicity, epigenetic targets for CAR T-cell exhaustion, and the role of the tumor microenvironment in driving relapse after CAR T therapy.

TReg cells serve as a biomarker for reduced CAR T neurotoxicity

Balance is the key to CAR T therapies, and clinical oncologists strive to enhance CAR T anti-tumor efficacy while reducing toxicity. Neurotoxicity is a particularly dangerous potential side effect of CAR T therapy, causing seizures, cerebral edema—swelling in the brain—and death (6). According to a 2021 report, the incidence of fatal neurotoxicity following anti-CD19 CAR T-cell therapy is 3%, and the incidence of severe neurotoxicity in a series of trials for large B-cell lymphoma (LBCL) ranged from 10% to 28% (7). Despite its prevalence, there are still few clear signs or clinical predictors of which patients will experience neurotoxicity, and to what degree of severity (6).

This motivated a team of scientists, led by Dr. Zinaida Good and Dr. Jay Spiegel from the Center for Cancer Cell Therapy and the Parker Institute for Cancer Immunotherapy at Stanford University School of Medicine, to search for biomarkers of disease progression and neurotoxicity in a cohort of 9 LBCL patients treated with a commercially available CD19-CAR, Axicabtagene ciloleucel (axi-cel) (6). Taking blood samples from patients at day 7 following axi-cel infusion, they sorted out CAR T cells then performed single cell immune profiling, including T-cell receptor (TCR) sequencing and cell surface protein profiling with Feature Barcoding technology, on the CAR product.

With this comprehensive immunophenotyping data, they identified three populations of CAR T cells: CD4+ and CD8+ populations of CD57+T-bet+ CAR T cells—both subsets were clonally expanded and showed high cytotoxic potential—and a CD4+CD57-Helios+ CAR T population that exhibited classical expression signatures associated with TReg cells (6). This latter population was mostly polyclonal, had low cytotoxicity potential, and, in a correlative study of 31 additional LBCL patients, was shown to be associated with poor disease control, but lowered risk for severe neurotoxicity. Their data also showed an inverse relationship between the CAR TReg population and CAR T-cell expansion, leading the team to conclude that the TReg population played a suppressive function to diminish anti-tumor response but also, as a byproduct, reduce toxicity. These findings may represent a catch-22: which is the lesser of two evils, continued disease progression or risk of neurotoxicity? Whatever the answer, this study ultimately provides a clinical biomarker to predict neurotoxicity, as well as a T-cell characteristic to target for modulation in future therapies.

An epigenetic answer to CAR T exhaustion

After a CAR T cell encounters cancer, the clock starts ticking—how long will the CAR maintain cytotoxic effector functions after contact with a cancer antigen? What changes take place in a CAR T cell that sustains or impairs its cancer-killing functions? The answers have huge importance for understanding the factors that influence CAR T-cell efficacy and persistence, which is ultimately crucial for disease outcomes (remission or relapse).

A research team from the First Affiliated Hospital, Zhejiang University School of Medicine in Hangzhou, China, sought to answer these questions, specifically investigating the role of gene regulation in driving CAR T-cell exhaustion (8). Using Single Cell ATAC (Assay for Transposase Accessible Chromatin), they studied the changes in chromatin accessibility that occurred in CAR T cells co-cultured with tumor cells. This revealed 12 distinct CAR T subtypes, including a cluster of terminally differentiated CD8+ CAR T cells that showed reduced chromatin accessibility for effector genes like IFNG and GZMB, and increased accessibility for genes that encoded immune inhibition receptors (such as LAG-3) and exhaustion-related transcription factors (8). Functional testing of LAG-3+ CAR T cells confirmed their reduced cytotoxicity, and that the ATAC data likely did reveal an exhausted CAR T-cell population.

The team then performed single cell ATAC-seq on clinical CAR T cells isolated at peak expansion and declining expansion stages from two multiple myeloma patients who had received BCMA CAR T therapy. They identified 7 CAR subtypes, including an exhausted CD8+ T-cell cluster that showed increased chromatin accessibility for the same exhaustion-specific transcription factors, such as BATF, IRF4, and PRDM1, from in vitro samples (8). This finding led the team to knock down BATF and/or IRF4 in another group of CAR T cells, then assess the effects on proliferation and cytotoxicity, both in vitro and in a mouse tumor model. Overall proliferation was not affected, but there was a change in the proportion of specific CAR T-cell subtypes after knockdown: IRF4 knockdown increased CAR T cells at the central memory stage, while BATF knockdown increased the proportion of naive CAR T cells (8). CAR T cells also showed reduced PD-1 expression and increased IFN production. Finally, in the mouse model, CAR T cells with BATF and IRF4 knockdown had better anti-tumor effects and drove prolonged mouse survival compared to controls (8).

These findings establish the importance of surveying the epigenetic regulatory landscape within individual CAR T cells, and point to potential epigenetic targets to overcome exhaustion and enhance therapeutic efficacy.

Cancer and the tumor microenvironment can sabotage CAR T-cell therapy

How effective a CAR T-cell therapy is may not always come down to the CAR product itself. There are a couple more variables in the mix—namely, the cancer and its surrounding cellular microenvironment.

Cancer cells have several resistance mechanisms in their toolkit, such as loss of the CAR T target antigen, upregulated expression of immune inhibitory receptors like PD-L1, or even dysfunctional apoptosis pathways that resist CAR T-cell killing (9). These tumor-intrinsic factors suggest that a multiomic, single cell characterization of the cancer itself may be essential foundational information to anticipate resistance and give CAR T-cell therapies a better fighting chance. The same can be said for the composition of the tumor microenvironment (TME), which may contain tissue barriers or specific cell types that secrete or express molecules that interfere with the activity and infiltration of immune effector cells (10). Understanding the complex role the TME may play in treatment resistance is all the more urgent as scientists press on towards advanced CAR T-cell therapies that can effectively treat both hematological and more challenging solid tumors.

A team of researchers from the University of Texas MD Anderson Cancer Center recently took a deep dive into the tumor-intrinsic and -extrinsic factors that affect CAR T-cell efficacy in mantle cell lymphoma, a rare, aggressive form of non-Hodgkin B-cell lymphoma (11). Their study utilized 39 longitudinal primary patient samples taken from peripheral blood, bone marrow, and excisional biopsies of the lymph node or spleen in 15 patients that had all failed a previous monoclonal antibody therapy—notably, this represents the largest longitudinal sampling to date for this cancer type (11). The samples, taken both pre- and post-Brexucabtagene autoleucel (BA) CAR T-cell therapy, were then profiled using single cell immune profiling and TCR sequencing.

Focusing their investigation on the tumor and TME, the team characterized 10 major immune lineages in TME cells, including CD8+ and CD4+ cytotoxic T cells (CTLs), monocytes, and natural killer (NK) cells. Next, the team wanted to identify cellular characteristics of the post-therapy TME in the context of relapse: single cell analysis of these samples revealed a trend towards decreased lymphoid cells, like CD8+ and CD4+ CTLs, and increased myeloid cells after relapse (11). Importantly, they also noted increased expression of TIGIT, an immune inhibitory checkpoint molecule that works to suppress the cytotoxic functions of both T and NK cells, on both mantle cell lymphoma tumor cells and CTLs after relapse. This acquired expression may represent the core mechanism leading to relapse after CAR T-cell therapy, and suggests co-targeting TIGIT could boost CAR T efficacy and improve patient outcomes (11).

Translating the research movement to improved CAR T therapies

As new findings about the underlying mechanisms of CAR T exhaustion, neurotoxicity, and therapeutic resistance build upon each other, the field presses on to ever-improving therapeutic strategies and the hope of better patient outcomes across cancer types. From improved quality of life, to longer periods of remission, every little step is a victory for the heroic patients who enlist in clinical trials and provide their own cells for study—and for future patients who can benefit from the findings carefully stewarded from those samples.

Advances in single cell multiomics serve to accelerate the path to that end goal. Other technology developments, including spatial assays, also continue to build a more comprehensive understanding of T cell–based cancer immunotherapies and the tumor response in the tissue context, providing crucial insights into cell-to-cell interactions and intercellular signaling that may be the key to unleashing CAR T therapies in solid tumors.

Continue the momentum by exploring how other scientists have leveraged single cell and spatial technology to pursue more effective CAR T therapies, including work led by Dr. Carl June and Dr. J. Joseph Melenhorst from the University of Pennsylvania exploring the secret to 10-year remission after CAR T therapy in advanced chronic lymphocytic leukemia. And read personal testimonials from Dr. Sneha Ramakrishna and Dr. Zinaida Good, who lead ongoing CAR T cell clinical trials for a fatal pediatric brain cancer at Stanford University School of Medicine.


  1. Yang J, et al. Advancing CAR T cell therapy through the use of multidimensional omics data. Nat Rev Clin Oncol 20: 211–228 (2023). doi: 10.1038/s41571-023-00729-2
  2. Kirouac D, et al. Deconvolution of clinical variance in CAR-T cell pharmacology and response. Nat Biotechnol (2023). doi: 10.1038/s41587-023-01687-x
  3. Majzner RG, et al. GD2-CAR T cell therapy for H3K27M-mutated diffuse midline gliomas. Nature 603: 934–941 (2022). doi: 10.1038/s41586-022-04489-4
  4. Kwon J, et al. Single-cell mapping of combinatorial target antigens for CAR switches using logic gates. Nat Biotechnol (2023). doi: 10.1038/s41587-023-01686-y
  5. Bucktrout S, et al. Advancing T cell–based cancer therapy with single-cell technologies. Nat Med 28: 1761–1764 (2022). doi: 10.1038/s41591-022-01986-x
  6. Good Z, et al. Post-infusion CAR TReg cells identify patients resistant to CD19-CAR therapy. Nat Med 28: 1860–1871 (2022). doi: 10.1038/s41591-022-01960-7
  7. Castaneda-Puglianini O and Chavez J. Assessing and management of neurotoxicity after CAR-T therapy in diffuse large B-cell lymphoma. J Blood Med 12: 775–783 (2021). doi: 10.2147/JBM.S281247
  8. Jiang P, et al. Single-cell ATAC-seq maps the comprehensive and dynamic chromatin accessibility landscape of CAR-T cell dysfunction. Leukemia 36: 2656–2668 (2022). doi: 10.1038/s41375-022-01676-0
  9. Lemoine J, et al. Born to survive: how cancer cells resist CAR T cell therapy. J Hematol Oncol 14: 199 (2021). doi: 10.1186/s13045-021-01209-9
  10. Cheng J, et al. Understanding the mechanisms of resistance to CAR T-cell therapy in malignancies. Front Oncol 9:1237 (2019). doi: 10.3389/fonc.2019.01237
  11. Jiang V, et al. TIGIT is the central player in T‑cell suppression associated with CAR T‑cell relapse in mantle cell lymphoma. Mol Cancer 21: 185 (2022). doi: 10.1186/s12943-022-01655-0